WO2021022900A1 - 识别文本的方法及装置 - Google Patents

识别文本的方法及装置 Download PDF

Info

Publication number
WO2021022900A1
WO2021022900A1 PCT/CN2020/095510 CN2020095510W WO2021022900A1 WO 2021022900 A1 WO2021022900 A1 WO 2021022900A1 CN 2020095510 W CN2020095510 W CN 2020095510W WO 2021022900 A1 WO2021022900 A1 WO 2021022900A1
Authority
WO
WIPO (PCT)
Prior art keywords
text
user
relationship
coefficient
importance
Prior art date
Application number
PCT/CN2020/095510
Other languages
English (en)
French (fr)
Inventor
宋增猛
龚嘉
谢月飞
王俊
汤华
马占寅
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP20849148.0A priority Critical patent/EP3992818A4/en
Publication of WO2021022900A1 publication Critical patent/WO2021022900A1/zh
Priority to US17/587,498 priority patent/US20220156294A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/40—Document-oriented image-based pattern recognition
    • G06V30/41—Analysis of document content
    • G06V30/413—Classification of content, e.g. text, photographs or tables
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G06F16/3325—Reformulation based on results of preceding query
    • G06F16/3326—Reformulation based on results of preceding query using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/3332—Query translation
    • G06F16/3334—Selection or weighting of terms from queries, including natural language queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • G06F16/353—Clustering; Classification into predefined classes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/107—Computer-aided management of electronic mailing [e-mailing]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/21—Monitoring or handling of messages
    • H04L51/226—Delivery according to priorities

Definitions

  • This application relates to the field of information classification, in particular to methods and devices for text recognition.
  • the text for example: instant chat content, emails, meeting notices, etc.
  • the text received or sent every day
  • the user needs to manually filter the text related to himself after receiving or sending the text.
  • manually filtering the text related to itself is not only time-consuming and inefficient.
  • the embodiments of the present application provide a method and device for recognizing text, which can recognize the importance of text received or sent by a user.
  • an embodiment of the present application provides a method for recognizing text.
  • the method includes: acquiring text and portrait information of a user, where the text is text received or sent by the user, and the portrait information of the user is used to indicate N keywords in the text related to the user, N is an integer greater than or equal to 1; the content feature coefficient is obtained according to the text and the portrait information of the user, wherein the content feature coefficient is used to indicate the importance of the content of the text Sex; Determine the importance of the text according to the content feature coefficient.
  • the technical solution provided by the above first aspect can obtain text and user portrait information, obtain content feature coefficients according to the text and user portrait information, and determine the importance of the text according to the content feature coefficients, thereby identifying users based on the content of the text The importance of the text received or sent.
  • the portrait information of the user includes N tag information, and the N tag information correspondingly indicate the N keywords; or, the portrait information of the user includes the N tag information and the N tags
  • the weight of each tag information in the message can be obtained according to the text and N tag information, and the importance of the text can be determined according to the content feature coefficients, or it can be based on the text, N tag information, and the value of each tag information in the N tag information.
  • the weight obtains the content feature coefficient, and determines the importance of the text according to the content feature coefficient, so that the importance of the text received or sent by the user can be recognized according to the content of the text.
  • the content feature coefficient includes N coefficients, where the nth coefficient is obtained according to the nth label information in the text and the portrait information of the user, and n is a positive value greater than 0 and less than or equal to N. Integer.
  • the nth coefficient in the content feature coefficients can be obtained according to the nth label information in the text and user portrait information, and the importance of the text can be determined according to the content feature coefficients, so that the user can identify whether the user received or received according to the content of the text. Importance of the text sent.
  • a possible implementation manner is to obtain the content feature coefficients according to the text and the portrait information of the user, including: obtaining first user behavior feedback data, where the first user behavior feedback data is used to indicate that the content of the text is historically similar The importance of the text with a high degree of importance; obtain a first correspondence, where the first correspondence is used to indicate the correlation between the first user behavior feedback data and the influence coefficient of the first user behavior feedback data; according to the The first user behavior feedback data and the first correspondence relationship obtain the influence coefficient of the first user behavior feedback data; the content characteristic coefficient is obtained according to the text, the portrait information of the user, and the influence coefficient of the first user behavior feedback data.
  • the first user behavior feedback data can be obtained, the influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data and the first relationship, and the influence coefficient of the first user behavior feedback data can be obtained according to the text, the portrait information of the user, and the first user behavior
  • the influence coefficient of the feedback data obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient, so that the importance of the text received or sent by the user can be recognized according to the content of the text.
  • the method further includes: obtaining user relationship information, where the user relationship information is used to indicate the hierarchical relationship between the user and other users; obtaining relationship feature coefficients according to the text and the user relationship information, Wherein, the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text; determining the importance of the text further includes: determining the importance of the text according to the relationship feature coefficient .
  • the relationship between the recipient and the sender identifies the importance of the text received or sent by the user.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user's subordinate relationship information, the user's department relationship information, the collaboration relationship information, and the user's communication relationship information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the frequency of communication between the user and other users, and the content feature coefficient and relationship
  • the feature coefficient determines the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the relationship feature coefficient can be determined according to the content feature coefficient and the relationship feature coefficient
  • the importance of the text can be used to identify the importance of the text received or sent by the user based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the subordinate relationship feature coefficient according to the user's subordinate relationship information obtains the department relationship feature coefficient according to the user's department relationship information, obtain the synergy relationship feature coefficient according to the collaboration relationship information, and obtain the communication based on the user's communication relationship information Relationship feature coefficient, and determine the importance of the text based on the content feature coefficient, the subordinate relationship feature coefficient, the department relationship feature coefficient, the collaboration relationship feature coefficient, and the communication relationship feature coefficient, so that the text content, the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the relationship characteristic coefficients include organizational relationship characteristic coefficients and communication relationship characteristic coefficients.
  • the organizational relationship characteristic coefficients include subordinate relationship characteristic coefficients, department relationship characteristic coefficients, and synergy relationship characteristic coefficients; according to the text and the Obtaining the relationship feature coefficient from the user's relationship information includes: obtaining second user behavior feedback data, where the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history; obtaining the second correspondence , Wherein the second correspondence is used to indicate the association relationship between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; according to the second user behavior feedback data and the second correspondence, Obtain the influence coefficient of the second user behavior feedback data; obtain the relationship characteristic coefficient according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence.
  • the influence coefficient of the user behavior feedback data obtains the relation characteristic coefficient, and the importance of the text is determined according to the content characteristic coefficient and the relation characteristic coefficient, so that the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text Or the importance of the text sent.
  • the text includes the identifier of the recipient of the text
  • the method further includes: obtaining the message influence range characteristic coefficient according to the identifier of the recipient of the text, wherein the message influence range characteristic coefficient is used to indicate The number of recipients of the text; and determining the importance of the text further includes: determining the importance of the text according to the characteristic coefficient of the influence range of the message.
  • the message influence range characteristic coefficient can be obtained according to the text, and the importance of the text can be determined according to the content characteristic coefficient, the relationship characteristic coefficient and the message influence range characteristic coefficient, so that the content of the text, the user and the recipient of the text can be determined
  • the relationship with the sender and the number of recipients of the text identify the importance of the text received or sent by the user.
  • the method further includes: obtaining the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, wherein the relationship coefficient between the user and the receiver/sender is used to indicate the relationship between the user and the The closeness of the relationship between the recipient of the text and/or the sender of the text; determining the importance of the text further includes: determining the importance of the text according to the relationship coefficient between the user and the recipient/sender.
  • the relationship coefficient between the user and the receiver/sender can be obtained according to the relationship information of the user, and the importance of the text can be determined according to the content feature coefficient, the relationship feature coefficient, and the relationship coefficient between the user and the receiver/sender.
  • the importance of the text received or sent by the user is recognized according to the content of the text, the relationship between the user and the recipient and sender of the text, and the closeness of the relationship between the user and the recipient and/or the sender of the text.
  • a possible implementation manner is that determining the importance of the text according to the content feature coefficient includes: using the content feature coefficient as input data of a machine learning method, and determining the importance of the text through the machine learning method. Based on the above technical solution, after the content feature coefficient is obtained, the content feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the determining the importance of the text according to the content feature coefficients includes: using the content feature coefficients and the relationship feature coefficients as input data of a machine learning method, and determining the importance of the text through the machine learning method . Based on the above technical solution, after the content feature coefficients and relationship feature coefficients are obtained, the content feature coefficients and relationship feature coefficients can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the user can recognize that the user receives or Importance of the text sent.
  • the determining the importance of the text according to the content feature coefficient includes: using the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient as the input data of the machine learning method, and the machine learning Method to determine the importance of the text.
  • the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient can be used as the input data of the machine learning method , Determine the importance of the text through machine learning methods, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficients, including: using the content feature coefficients, the relationship feature coefficients, the message influence range feature coefficients, and the relationship coefficients between the user and the recipient/sender As the input data of the machine learning method, the importance of the text is determined by the machine learning method.
  • the content feature coefficients, relationship feature coefficients, message influence range feature coefficients, and user-recipient/sender relationship coefficients is used as the input data of the machine learning method, and the importance of the text is determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the content feature coefficient when there are multiple texts of the same importance, the content feature coefficient includes N coefficients, and the method further includes: adding all or part of the N coefficients in the content feature coefficient to obtain the The importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • all or part of the N coefficients in the content feature coefficients can be added to obtain the importance coefficient of the text, and according to the text
  • the importance coefficient ranks the text received or sent by the user, which enables the user to process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, the content feature coefficient includes N coefficients, and the method further includes: the sum of the coefficients in the content feature coefficients and the coefficient in the relationship feature coefficients Multiplication is performed to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to obtain the importance of the text. It also sorts the text received or sent by the user according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, includes N coefficients, and the method further includes: the sum of the coefficients in the content feature coefficients and the coefficient in the relationship feature coefficients Do a multiplication operation, and perform a division operation with the characteristic coefficient of the message's influence range to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to affect the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, includes N coefficients.
  • the method further includes: the sum of the coefficients in the content feature coefficients and the relation feature coefficients The coefficient and the coefficient of the relationship between the user and the recipient/sender are multiplied, and then divided by the characteristic coefficient of the influence range of the message to obtain the importance coefficient of the text; the text received or sent by the user according to the importance coefficient of the text put in order.
  • the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the user and recipient/ The relationship coefficient of the sender is multiplied, and the feature coefficient of the influence range of the message is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text. Importance of processing text.
  • a possible implementation manner is to determine the importance of the text according to the content feature coefficients, including: adding all or part of the N coefficients in the content feature coefficients to obtain the importance coefficient of the text; according to the importance of the text
  • the gender factor determines the importance of the text.
  • a possible implementation manner is that determining the importance of the text according to the content feature coefficients includes: multiplying the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients to obtain the importance of the text; The importance coefficient of the text determines the importance of the text. Based on the above technical solution, after the content feature coefficient and the relationship feature coefficient are obtained, the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied to obtain the importance coefficient of the text. The importance coefficient of determines the importance of the text, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficients, including: multiplying the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients, and the message influence range feature coefficients Do the division operation to get the importance coefficient of the text; determine the importance of the text according to the importance coefficient of the text.
  • the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied by The feature coefficient of the message influence range is divided to obtain the importance coefficient of the text, and the importance of the text is determined according to the importance coefficient of the text, so that the importance of the text received or sent by the user can be recognized.
  • determining the importance of the text according to the content feature coefficients includes: the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the relationship between the user and the recipient/sender The relationship coefficient is multiplied and divided with the characteristic coefficient of the influence range of the message to obtain the importance coefficient of the text; the importance coefficient of the text is determined according to the importance coefficient of the text.
  • the sum of the content characteristic coefficients can be combined with The coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender are multiplied, and the message influence range feature coefficient is divided by the coefficient of the text to obtain the importance coefficient of the text, and the text is determined according to the importance coefficient of the text.
  • the importance of the text which can identify the importance of the text received or sent by the user.
  • the method further includes: sorting the texts received or sent by the user according to the importance coefficient of the texts. Based on the above technical solution, after determining the importance of the text, when there are multiple texts with the same importance, the text received or sent by the user can be sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text .
  • the method further includes: classifying and displaying the text according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • an embodiment of the present application provides a method for recognizing text.
  • the method includes: obtaining text and user relationship information, where the text is a text received or sent by the user, and the user relationship information is used to indicate the The hierarchical relationship between the user and other users; the relationship feature coefficient is obtained according to the text and the relationship information of the user, where the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text; Determine the importance of the text according to the relationship feature coefficient.
  • the technical solution provided by the above second aspect can obtain the relationship information of the text and the user, obtain the relationship feature coefficient according to the relationship information of the text and the user, and determine the importance of the text according to the relationship feature coefficient, so that the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user’s superior and subordinate relationship information, the user’s department relationship information, the collaboration relationship information, and the user’s communication relationship information, and the importance of the text can be determined according to the relationship feature coefficient.
  • the relationship between the user and the recipient and sender of the text identifies the importance of the text received or sent by the user.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the communication frequency between the user and other users, and the text can be determined according to the relationship feature coefficient Therefore, the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the importance of the text can be determined according to the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the relationship characteristic coefficients include organizational relationship characteristic coefficients and communication relationship characteristic coefficients.
  • the organizational relationship characteristic coefficients include subordinate relationship characteristic coefficients, department relationship characteristic coefficients, and synergy relationship characteristic coefficients; according to the text and the Obtaining the relationship feature coefficient from the user's relationship information includes: obtaining second user behavior feedback data, where the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history; obtaining the second correspondence , Wherein the second correspondence is used to indicate the association relationship between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; according to the second user behavior feedback data and the second correspondence, Obtain the influence coefficient of the second user behavior feedback data; obtain the relationship characteristic coefficient according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence.
  • the text the user's relationship information, and the first 2.
  • the influence coefficient of the user behavior feedback data obtains the relationship feature coefficient, and the importance of the text is determined according to the relationship feature coefficient, so that the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • a possible implementation manner is that determining the importance of the text according to the relationship feature coefficient includes: using the relationship feature coefficient as input data of a machine learning method, and determining the importance of the text through the machine learning method. Based on the above technical solution, after the relationship feature coefficient is obtained, the relationship feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the method further includes: multiplying the coefficients in the relation feature coefficients to obtain the importance coefficient coefficient of the text; according to the importance coefficient of the text Sort the text received or sent by the user.
  • the coefficient in the relation feature coefficient can be multiplied to obtain the importance coefficient of the text, and the user can be Sorting the received or sent text allows users to process the text according to the importance of the text.
  • determining the importance of the text according to the relationship feature coefficient includes: multiplying the coefficients in the relationship feature coefficient to obtain the importance coefficient of the text; and determining the importance coefficient of the text according to the importance coefficient of the text. The importance of text. Based on the above technical solutions, after the relationship feature coefficients are obtained, the coefficients in the relationship feature coefficients can be multiplied to obtain the importance coefficient of the text, and the importance of the text can be determined according to the importance coefficient of the text, so as to identify the user receiving or sending The importance of the text.
  • the method further includes: sorting the texts received or sent by the user according to the importance coefficient of the texts. Based on the above technical solution, after determining the importance of the text, when there are multiple texts with the same importance, the text received or sent by the user can be sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text .
  • the method further includes: classifying and displaying the text according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • an embodiment of the present application provides a method for recognizing text.
  • the method includes: obtaining text, user portrait information, and user relationship information, where the text is the text received or sent by the user, and the user portrait
  • the information is used to indicate N keywords in the text related to the user, where N is an integer greater than or equal to 1, and the user’s relationship information is used to indicate the hierarchical relationship between the user and other users; according to the text and the user’s
  • the portrait information obtains content feature coefficients, where the content feature coefficients are used to indicate the importance of the content of the text; the relationship feature coefficients are obtained according to the text and the relationship information of the user, where the relationship feature coefficients are used to indicate the relationship between the user and the user.
  • the relationship between the recipient of the text and/or the sender of the text; the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient.
  • the technical solution provided by the above third aspect can obtain text, user portrait information, and user relationship information, obtain content feature coefficients based on text and user portrait information, obtain relationship feature coefficients based on text and user relationship information, and obtain relationship feature coefficients based on content
  • the feature coefficient and the relationship feature coefficient determine the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the portrait information of the user includes N tag information, and the N tag information correspondingly indicate the N keywords; or, the portrait information of the user includes the N tag information and the N tags
  • the weight of each tag information in the message can be obtained according to the text and N tag information, and the importance of the text can be determined according to the content feature coefficients and the relationship feature coefficients, or the text, N tag information, and each of the N tag information
  • the weight of each tag information obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient, so that the content received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text The importance of text.
  • the content feature coefficient includes N coefficients, where the nth coefficient is obtained according to the nth label information in the text and the portrait information of the user, and n is a positive value greater than 0 and less than or equal to N. Integer.
  • the nth coefficient in the content feature coefficient can be obtained according to the nth label information in the text and user portrait information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient, so that the content of the text can be determined , And the relationship between the user and the recipient and sender of the text to identify the importance of the text received or sent by the user.
  • a possible implementation manner is to obtain content feature coefficients according to the text and portrait information of the user, including: obtaining first user behavior feedback data, where the first user behavior feedback data is used to indicate the historical similarity to the text content High text importance; acquiring a first corresponding relationship, where the first corresponding relationship is used to indicate the association relationship between the first user behavior feedback data and the influence coefficient of the first user behavior feedback data; according to the first A user behavior feedback data and the first correspondence relationship obtain the influence coefficient of the first user behavior feedback data; the content characteristic coefficient is obtained according to the text, the portrait information of the user, and the influence coefficient of the first user behavior feedback data.
  • the first user behavior feedback data can be obtained, the influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data and the first relationship, and the influence coefficient of the first user behavior feedback data can be obtained according to the text, the portrait information of the user, and the first user behavior
  • the influence coefficient of the feedback data obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient, so that the content received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text The importance of text.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user's subordinate relationship information, the user's department relationship information, the collaboration relationship information, and the user's communication relationship information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the frequency of communication between the user and other users, and the content feature coefficient and relationship
  • the feature coefficient determines the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the relationship feature coefficient can be determined according to the content feature coefficient and the relationship feature coefficient
  • the importance of the text can be used to identify the importance of the text received or sent by the user based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the subordinate relationship feature coefficient according to the user's subordinate relationship information obtains the department relationship feature coefficient according to the user's department relationship information, obtain the synergy relationship feature coefficient according to the collaboration relationship information, and obtain the communication based on the user's communication relationship information Relationship feature coefficient, and determine the importance of the text based on the content feature coefficient, the subordinate relationship feature coefficient, the department relationship feature coefficient, the collaboration relationship feature coefficient, and the communication relationship feature coefficient, so that the text content, the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the relationship characteristic coefficients include organizational relationship characteristic coefficients and communication relationship characteristic coefficients.
  • the organizational relationship characteristic coefficients include subordinate relationship characteristic coefficients, department relationship characteristic coefficients, and synergy relationship characteristic coefficients; according to the text and the Obtaining the relationship feature coefficient from the user's relationship information includes: obtaining second user behavior feedback data, where the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history; obtaining the second correspondence , Wherein the second correspondence is used to indicate the association relationship between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; according to the second user behavior feedback data and the second correspondence, Obtain the influence coefficient of the second user behavior feedback data; obtain the relationship characteristic coefficient according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence.
  • the influence coefficient of the user behavior feedback data obtains the relation characteristic coefficient, and the importance of the text is determined according to the content characteristic coefficient and the relation characteristic coefficient, so that the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text Or the importance of the text sent.
  • the method further includes: obtaining a characteristic coefficient of the influence range of the message according to the text, wherein the characteristic coefficient of the influence range of the message is used to indicate the number of recipients of the text; and determining the importance of the text , And also includes: determining the importance of the text according to the characteristic coefficient of the influence range of the message.
  • the message influence range characteristic coefficient can be obtained according to the text, and the importance of the text can be determined according to the content characteristic coefficient, the relationship characteristic coefficient and the message influence range characteristic coefficient, so that the content of the text, the user and the recipient of the text can be determined
  • the relationship of the sender and the number of recipients of the text identify the importance of the text received or sent by the user.
  • the method further includes: obtaining the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, wherein the relationship coefficient between the user and the receiver/sender is used to indicate that the user is The degree of closeness of the relationship between the recipient of the text and/or the sender of the text; determining the importance of the text further includes: determining the importance of the text according to the relationship coefficient between the user and the recipient/sender.
  • the relationship coefficient between the user and the recipient/sender can be obtained according to the relationship information of the user, and the importance of the text can be determined according to the content feature coefficient, the relationship feature coefficient, and the relationship coefficient between the user and the recipient/sender.
  • the importance of the text received or sent by the user is recognized according to the content of the text, the relationship between the user and the recipient and sender of the text, and the closeness of the relationship between the user and the recipient and/or the sender of the text.
  • the determining the importance of the text according to the content feature coefficient and the relation feature coefficient includes: using the content feature coefficient and the relation feature coefficient as input data of a machine learning method, and determining the text by the machine learning method The importance of text. Based on the above technical solution, after the feature coefficient set is obtained, the content feature coefficients and relationship feature coefficients can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the text received or sent by the user can be identified importance.
  • the determining the importance of the text according to the content feature coefficient and the relationship feature coefficient includes: using the content feature coefficient, the relationship feature coefficient and the message influence range feature coefficient as the input data of the machine learning method, Determine the importance of the text through the machine learning method.
  • the content feature coefficient, the relationship feature coefficient and the message influence range feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so as to identify The importance of the text received or sent by the user.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficient and the relationship feature coefficient, including: using the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the user and the recipient/sender
  • the human relationship coefficient is used as the input data of the machine learning method, and the importance of the text is determined by the machine learning method.
  • the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender can be used as the input data of the machine learning method.
  • the learning method determines the importance of the text, so that the importance of the text received or sent by the user can be recognized.
  • the content feature coefficient when there are multiple texts with the same importance, the content feature coefficient includes N coefficients, and the method further includes: the sum of the coefficients in the content feature coefficients and the coefficient in the relationship feature coefficients Multiplication is performed to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to obtain the importance of the text. It also sorts the text received or sent by the user according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, includes N coefficients, and the method further includes: the sum of the coefficients in the content feature coefficients and the coefficient in the relationship feature coefficients Do a multiplication operation, and perform a division operation with the characteristic coefficient of the message's influence range to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to affect the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, includes N coefficients.
  • the method further includes: the sum of the coefficients in the content feature coefficients and the relation feature coefficients The coefficient and the coefficient of the relationship between the user and the recipient/sender are multiplied, and then divided by the characteristic coefficient of the influence range of the message to obtain the importance coefficient of the text; the text received or sent by the user according to the importance coefficient of the text put in order.
  • the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the user and recipient/ The relationship coefficient of the sender is multiplied, and the feature coefficient of the influence range of the message is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text. Importance of processing text.
  • a possible implementation manner is to determine the importance of the text according to content feature coefficients and relation feature coefficients, including: multiplying the sum of the coefficients in the content feature coefficients and the coefficients in the relation feature coefficients to obtain the text Importance coefficient; the importance of the text is determined according to the importance coefficient of the text.
  • the content feature coefficient and the relationship feature coefficient are obtained, the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied to obtain the importance coefficient of the text.
  • the importance coefficient of determines the importance of the text, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation manner is to determine the importance of the text according to the content feature coefficients and the relation feature coefficients, including: multiplying the sum of the coefficients in the content feature coefficients with the coefficients in the relation feature coefficients, and the message
  • the influence range feature coefficient is divided to obtain the importance coefficient of the text; the importance of the text is determined according to the importance coefficient of the text.
  • the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied by
  • the feature coefficient of the message influence range is divided to obtain the importance coefficient of the text, and the importance of the text is determined according to the importance coefficient of the text, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficient and the relation feature coefficient, including: the sum of the coefficients in the content feature coefficient, the coefficient in the relation feature coefficient, and the user and the recipient
  • the relationship coefficient of the sender is multiplied and divided by the characteristic coefficient of the influence range of the message to obtain the importance coefficient of the text; the importance of the text is determined according to the importance coefficient of the text.
  • the sum of the content characteristic coefficients can be combined with The coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender are multiplied, and the message influence range feature coefficient is divided by the coefficient of the text to obtain the importance coefficient of the text, and the text is determined according to the importance coefficient of the text.
  • the importance of the text which can identify the importance of the text received or sent by the user.
  • the method further includes: sorting the texts received or sent by the user according to the importance coefficient of the texts. Based on the above technical solution, after determining the importance of the text, when there are multiple texts with the same importance, the text received or sent by the user can be sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text .
  • the method further includes: classifying and displaying the text according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • an embodiment of the present application provides an apparatus for recognizing text.
  • the apparatus includes: an acquisition module, a processing module, and a determination module; the acquisition module is used to acquire text and portrait information of a user, where the text is the user The text received or sent, the portrait information of the user is used to indicate the N keywords in the text related to the user, and N is an integer greater than or equal to 1; the processing module is used for the text and the portrait information of the user A content feature coefficient is obtained, where the content feature coefficient is used to indicate the importance of the content of the text; the determining module is used to determine the importance of the text according to the content feature coefficient.
  • the technical solution provided by the above-mentioned fourth aspect can obtain text and user portrait information, obtain content feature coefficients according to the text and user portrait information, and determine the importance of the text according to the content feature coefficients, thereby identifying users based on the content of the text The importance of the text received or sent.
  • the portrait information of the user includes N tag information, and the N tag information correspondingly indicate the N keywords; or, the portrait information of the user includes the N tag information and the N tags
  • the weight of each tag information in the message can be obtained according to the text and N tag information, and the importance of the text can be determined according to the content feature coefficients, or it can be based on the text, N tag information, and the value of each tag information in the N tag information.
  • the weight obtains the content feature coefficient, and determines the importance of the text according to the content feature coefficient, so that the importance of the text received or sent by the user can be recognized according to the content of the text.
  • the content feature coefficient includes N coefficients, where the nth coefficient is obtained according to the nth label information in the text and the portrait information of the user, and n is a positive value greater than 0 and less than or equal to N. Integer.
  • the nth coefficient in the content feature coefficients can be obtained according to the nth label information in the text and user portrait information, and the importance of the text can be determined according to the content feature coefficients, so that the user can identify whether the user receives or Importance of the text sent.
  • the processing module is specifically used to obtain the first user behavior feedback data, where the first user behavior feedback data is used to indicate the importance of a text with a high similarity to the text content in history; this processing The module is also specifically configured to obtain a first correspondence relationship, where the first correspondence relationship is used to indicate an association relationship between the first user behavior feedback data and the influence coefficient of the first user behavior feedback data; the processing module, It is also specifically configured to obtain the influence coefficient of the first user behavior feedback data according to the first user behavior feedback data and the first correspondence; the processing module is also specifically configured to obtain the influence coefficient of the first user behavior feedback data according to the text, the portrait information of the user, and the first correspondence.
  • An influence coefficient of user behavior feedback data obtains the content characteristic coefficient.
  • the first user behavior feedback data can be obtained, the influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data and the first relationship, and the influence coefficient of the first user behavior feedback data can be obtained according to the text, the portrait information of the user, and the first user behavior
  • the influence coefficient of the feedback data obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient, so that the importance of the text received or sent by the user can be recognized according to the content of the text.
  • the acquisition module is also used to acquire user relationship information, where the user relationship information is used to indicate the hierarchical relationship between the user and other users;
  • the processing module is also used to The relationship information of the user obtains the relationship feature coefficient, wherein the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text, and the determining module is also used to determine the relationship feature coefficient The importance of the text.
  • the relationship between the recipient and the sender identifies the importance of the text received or sent by the user.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user's subordinate relationship information, the user's department relationship information, the collaboration relationship information, and the user's communication relationship information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the frequency of communication between the user and other users, and the content feature coefficient and relationship
  • the feature coefficient determines the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the relationship feature coefficient can be determined according to the content feature coefficient and the relationship feature coefficient
  • the importance of the text can be used to identify the importance of the text received or sent by the user based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the subordinate relationship feature coefficient according to the user's subordinate relationship information obtains the department relationship feature coefficient according to the user's department relationship information, obtain the synergy relationship feature coefficient according to the collaboration relationship information, and obtain the communication based on the user's communication relationship information Relationship feature coefficient, and determine the importance of the text based on the content feature coefficient, the subordinate relationship feature coefficient, the department relationship feature coefficient, the collaboration relationship feature coefficient, and the communication relationship feature coefficient, so that the text content, the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the relationship feature coefficients include organizational relationship feature coefficients and communication relationship feature coefficients, and the organizational relationship feature coefficients include subordinate relationship feature coefficients, department relationship feature coefficients, and collaboration relationship feature coefficients;
  • the processing module specifically uses Used to obtain the second user behavior feedback data, where the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history;
  • the processing module is also specifically used to The behavior feedback data and the second correspondence relationship are used to obtain the influence coefficient of the second user behavior feedback data, where the second correspondence relationship is used to indicate the relationship between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data
  • the processing module is also specifically configured to obtain the relationship feature coefficient according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence.
  • the influence coefficient of the user behavior feedback data obtains the relation characteristic coefficient, and the importance of the text is determined according to the content characteristic coefficient and the relation characteristic coefficient, so that the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text Or the importance of the text sent.
  • the text includes the identifier of the recipient of the text
  • the processing module is further configured to obtain the message influence range characteristic coefficient according to the identifier of the recipient of the text, wherein the message influence range characteristic coefficient is
  • the determining module is also used to determine the importance of the text according to the characteristic coefficient of the influence range of the message.
  • the message influence range characteristic coefficient can be obtained according to the text, and the importance of the text can be determined according to the content characteristic coefficient, the relationship characteristic coefficient and the message influence range characteristic coefficient, so that the content of the text, the user and the recipient of the text can be determined
  • the relationship with the sender and the number of recipients of the text identify the importance of the text received or sent by the user.
  • the processing module is also used to obtain the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, wherein the relationship coefficient between the user and the receiver/sender is used to indicate the user
  • the determining module is also used to determine the importance of the text according to the relationship coefficient between the user and the recipient/sender.
  • the relationship coefficient between the user and the receiver/sender can be obtained according to the relationship information of the user, and the importance of the text can be determined according to the content feature coefficient, the relationship feature coefficient, and the relationship coefficient between the user and the receiver/sender.
  • the importance of the text received or sent by the user is recognized according to the content of the text, the relationship between the user and the recipient and sender of the text, and the closeness of the relationship between the user and the recipient and/or the sender of the text.
  • the determination module is specifically configured to use the content feature coefficients as input data of a machine learning method, and determine the importance of the text through the machine learning method. Based on the above technical solution, after the content feature coefficient is obtained, the content feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized .
  • the determining module is specifically configured to use the content feature coefficient and the relationship feature coefficient as input data of a machine learning method, and determine the importance of the text through the machine learning method. Based on the above technical solution, after the content feature coefficients and relationship feature coefficients are obtained, the content feature coefficients and relationship feature coefficients can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the user can recognize that the user receives or Importance of the text sent.
  • the determining module is specifically configured to use the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient as the input data of the machine learning method, and the importance of the text is determined by the machine learning method .
  • the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient can be used as the input data of the machine learning method , Determine the importance of the text through machine learning methods, so that the importance of the text received or sent by the user can be recognized.
  • the determining module is specifically configured to use the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender as the input data of the machine learning method , To determine the importance of the text through the machine learning method.
  • the content feature coefficients, relationship feature coefficients, message influence range feature coefficients, and user-recipient/sender relationship coefficients After obtaining the content feature coefficients, relationship feature coefficients, message influence range feature coefficients, and user-recipient/sender relationship coefficients, the content feature coefficients, relationship feature coefficients, message influence range feature coefficients, and user
  • the relationship coefficient with the recipient/sender is used as the input data of the machine learning method, and the importance of the text is determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the content feature coefficient when there are multiple texts with the same importance, includes N coefficients, and the device further includes: a sorting module; the processing module is further used for N of the content feature coefficients. All or part of the coefficients are added to obtain the importance coefficient coefficient of the text; the sorting module is used to sort the text received or sent by the user according to the importance coefficient of the text.
  • the device further includes: a sorting module; the processing module is further used for N of the content feature coefficients. All or part of the coefficients are added to obtain the importance coefficient coefficient of the text; the sorting module is used to sort the text received or sent by the user according to the importance coefficient of the text.
  • the device further includes: a sorting module; the processing module is further used for the coefficient in the content feature coefficient The sum and the coefficient in the relationship feature coefficient are multiplied to obtain the importance coefficient of the text; the sorting module is used to sort the text received or sent by the user according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to obtain the importance of the text. It also sorts the text received or sent by the user according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient includes N coefficients
  • the device further includes: a sorting module; the processing module is further used for the coefficient in the content feature coefficient The sum is multiplied by the coefficient in the relationship feature coefficient, and the message is divided by the feature coefficient of the influence range of the message to obtain the importance coefficient of the text; the sorting module is used to receive or receive the user according to the importance coefficient of the text The sent text is sorted.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to affect the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, the content feature coefficient includes N coefficients, and the device further includes: a sorting module; the processing module is further used for the coefficient in the content feature coefficient The sum is multiplied by the coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender, and divided by the message influence range feature coefficient to obtain the importance coefficient of the text; the ranking module uses In order to sort the text received or sent by the user according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the user and recipient/ The relationship coefficient of the sender is multiplied, and the feature coefficient of the influence range of the message is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text. Importance of processing text.
  • the determining module is specifically used to add all or part of the N coefficients in the content feature coefficients to obtain the importance coefficient of the text; the determining module is also specifically used to determine the importance of the text according to The importance factor determines the importance of the text. Based on the above technical solution, after the content feature coefficients are obtained, all or part of the N coefficients in the content feature coefficients can be added to obtain the importance coefficient of the text, and the importance of the text can be determined according to the importance coefficient of the text , Which can identify the importance of the text received or sent by the user.
  • the determining module is specifically used to multiply the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; the determining module also specifically uses To determine the importance of the text according to the importance coefficient of the text. Based on the above technical solution, after the content feature coefficient and the relation feature coefficient are obtained, the sum of the coefficients in the content feature coefficient and the coefficient in the relation feature coefficient can be multiplied to obtain the importance coefficient of the text. Identify the importance of text received or sent by the user.
  • the determining module is specifically used to multiply the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients, and divide with the feature coefficients of the message influence range to obtain the text Importance coefficient; the determining module is also specifically used to determine the importance of the text according to the importance coefficient of the text.
  • the determination module is specifically configured to multiply the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the relationship coefficients between the user and the receiver/sender, and The feature coefficient of the influence range of the message is divided to obtain the importance coefficient of the text; the determining module is also specifically used to determine the importance of the text according to the importance coefficient of the text.
  • the sum of the content characteristic coefficients can be combined with The coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender are multiplied, and the message influence range feature coefficient is divided by the coefficient of importance of the text, so that the text received or sent by the user can be identified The importance of.
  • the device when there are multiple texts with the same importance, the device further includes: a sorting module; the sorting module is used to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the sorting module is used to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the device further includes: a display module; the display module is used to classify and display the text according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • an embodiment of the present application provides an apparatus for recognizing text.
  • the apparatus includes: an acquisition module, a processing module, and a determination module; the acquisition module is used to acquire text and user relationship information, where the text is the user In the text received or sent, the relationship information of the user is used to indicate the hierarchical relationship between the user and other users; the processing module is used to obtain the relationship feature coefficient according to the text and the relationship information of the user, wherein the relationship feature coefficient is To indicate the relationship between the user and the recipient of the text and/or the sender of the text; the determining module is used to determine the importance of the text according to the relationship feature coefficient.
  • the technical solution provided by the above fifth aspect can obtain the relationship information of the text and the user, obtain the relationship feature coefficient according to the relationship information of the text and the user, and determine the importance of the text according to the relationship feature coefficient, so that the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user’s superior and subordinate relationship information, the user’s department relationship information, the collaboration relationship information, and the user’s communication relationship information, and the importance of the text can be determined according to the relationship feature coefficient.
  • the relationship between the user and the recipient and sender of the text identifies the importance of the text received or sent by the user.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the communication frequency between the user and other users, and the text can be determined according to the relationship feature coefficient Therefore, the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the importance of the text can be determined according to the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the relationship feature coefficients include organizational relationship feature coefficients and communication relationship feature coefficients, and the organizational relationship feature coefficients include subordinate relationship feature coefficients, department relationship feature coefficients, and collaboration relationship feature coefficients;
  • the processing module specifically uses To obtain the second user behavior feedback data, the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history; the processing module is also specifically used to obtain the second correspondence, where , The second correspondence is used to indicate the second user behavior feedback data and the association relationship between the influence coefficients of the second user behavior feedback data; the processing module is also specifically used to feedback data according to the second user behavior And the second corresponding relationship to obtain the influence coefficient of the second user behavior feedback data; the processing module is further specifically configured to obtain the relationship according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data Characteristic coefficient.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence.
  • the text the user's relationship information, and the first 2.
  • the influence coefficient of the user behavior feedback data obtains the relationship feature coefficient, and the importance of the text is determined according to the relationship feature coefficient, so that the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • the determining module is specifically configured to use the relationship feature coefficient as the input data of the machine learning method, and determine the importance of the text through the machine learning method. Based on the above technical solution, after the relationship feature coefficient is obtained, the relationship feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the device when there are multiple texts with the same importance, the device further includes: a sorting module; the processing module is also used to multiply the coefficients in the relation feature coefficients to obtain the importance of the text Coefficient; the sorting module is used to sort the text received or sent by the user according to the importance coefficient of the text.
  • a sorting module is used to sort the text received or sent by the user according to the importance coefficient of the text.
  • the determining module is specifically used to multiply the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; the determining module is also specifically used to determine the importance coefficient of the text according to the importance coefficient of the text.
  • the importance of text Based on the above technical solutions, after the relationship feature coefficients are obtained, the coefficients in the relationship feature coefficients can be multiplied to obtain the importance coefficient of the text, and the importance of the text can be determined according to the importance coefficient of the text, so as to identify the user receiving or sending The importance of the text.
  • the device when there are multiple texts with the same importance, the device further includes: a sorting module; the sorting module is used to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the sorting module is used to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the device further includes: a display module; the display module is used to classify and display the text according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • an embodiment of the present application provides a text recognition device, which includes: an acquisition module, a processing module, and a determination module; the acquisition module is used to acquire text, user portrait information, and user relationship information, where , The text is the text received or sent by the user, the portrait information of the user is used to indicate the N keywords in the text related to the user, and N is an integer greater than or equal to 1, and the relationship information of the user is used to indicate the The hierarchical relationship between the user and other users; the processing module is used to obtain content feature coefficients based on the text and the user’s portrait information, where the content feature coefficients are used to indicate the importance of the content of the text; the processing module also uses The relationship feature coefficient is obtained according to the text and the relationship information of the user, where the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text; the determining module is used to The content feature coefficient and the relationship feature coefficient determine the importance of the text.
  • the technical solution provided by the above sixth aspect can obtain text, user portrait information, and user relationship information, obtain content feature coefficients based on text and user portrait information, obtain relationship feature coefficients based on text and user relationship information, and obtain relationship feature coefficients based on content
  • the feature coefficient and the relationship feature coefficient determine the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the portrait information of the user includes N tag information, and the N tag information correspondingly indicate the N keywords; or, the portrait information of the user includes the N tag information and the N tags
  • the weight of each tag information in the message can be obtained according to the text and N tag information, and the importance of the text can be determined according to the content feature coefficients and the relationship feature coefficients, or the text, N tag information, and each of the N tag information
  • the weight of each tag information obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient, so that the content received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text The importance of text.
  • the content feature coefficient includes N coefficients, where the nth coefficient is obtained according to the nth label information in the text and the portrait information of the user, and n is a positive value greater than 0 and less than or equal to N. Integer.
  • the nth coefficient in the content feature coefficient can be obtained according to the nth label information in the text and user portrait information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient, so that the content of the text can be determined , And the relationship between the user and the recipient and sender of the text to identify the importance of the text received or sent by the user.
  • the processing module is specifically used to obtain the first user behavior feedback data, where the first user behavior feedback data is used to indicate the importance of a text with a high similarity to the text content in history; this processing The module is also specifically configured to obtain a first correspondence relationship, where the first correspondence relationship is used to indicate an association relationship between the first user behavior feedback data and the influence coefficient of the first user behavior feedback data; the processing module, It is also specifically configured to obtain the influence coefficient of the first user behavior feedback data according to the first user behavior feedback data and the first correspondence; the processing module is also specifically configured to obtain the influence coefficient of the first user behavior feedback data according to the text, the portrait information of the user, and the first correspondence.
  • An influence coefficient of user behavior feedback data obtains the content characteristic coefficient.
  • the first user behavior feedback data can be obtained, the influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data and the first relationship, and the influence coefficient of the first user behavior feedback data can be obtained according to the text, the portrait information of the user, and the first user behavior
  • the influence coefficient of the feedback data obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient, so that the content received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text The importance of text.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user's subordinate relationship information, the user's department relationship information, the collaboration relationship information, and the user's communication relationship information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the frequency of communication between the user and other users, and the content feature coefficient and relationship
  • the feature coefficient determines the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the relationship feature coefficient can be determined according to the content feature coefficient and the relationship feature coefficient
  • the importance of the text so that the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the subordinate relationship feature coefficient according to the user's subordinate relationship information obtains the department relationship feature coefficient according to the user's department relationship information, obtain the synergy relationship feature coefficient according to the collaboration relationship information, and obtain the communication based on the user's communication relationship information Relationship feature coefficient, and determine the importance of the text based on the content feature coefficient, the subordinate relationship feature coefficient, the department relationship feature coefficient, the collaboration relationship feature coefficient, and the communication relationship feature coefficient, so that the text content, the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the relationship feature coefficients include organizational relationship feature coefficients and communication relationship feature coefficients, and the organizational relationship feature coefficients include subordinate relationship feature coefficients, department relationship feature coefficients, and collaboration relationship feature coefficients;
  • the processing module specifically uses To obtain the second user behavior feedback data, the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history; the processing module is also specifically used to obtain the second correspondence, where , The second correspondence relationship is used to indicate the association relationship between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; the processing module is also specifically used to feedback data according to the second user behavior and The second corresponding relationship is used to obtain the influence coefficient of the second user behavior feedback data; the processing module is further specifically configured to obtain the relationship feature according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data coefficient.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence.
  • the influence coefficient of the user behavior feedback data obtains the relation characteristic coefficient, and the importance of the text is determined according to the content characteristic coefficient and the relation characteristic coefficient, so that the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text Or the importance of the text sent.
  • the processing module is further configured to obtain a characteristic coefficient of the influence range of the message according to the text, wherein the characteristic coefficient of the influence range of the message is used to indicate the number of recipients of the text, and the determining module also It is used to determine the importance of the text according to the characteristic coefficient of the influence range of the message.
  • the message influence range characteristic coefficient can be obtained according to the text, and the importance of the text can be determined according to the content characteristic coefficient, the relationship characteristic coefficient and the message influence range characteristic coefficient, so that the content of the text, the user and the recipient of the text can be determined
  • the relationship of the sender and the number of recipients of the text identify the importance of the text received or sent by the user.
  • the processing module is also used to obtain the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, wherein the relationship coefficient between the user and the receiver/sender is used to indicate the The degree of closeness of the relationship between the user and the recipient of the text and/or the sender of the text
  • the determining module is also used to determine the importance of the text according to the relationship coefficient between the user and the recipient/sender.
  • the relationship coefficient between the user and the recipient/sender can be obtained according to the relationship information of the user, and the importance of the text can be determined according to the content feature coefficient, the relationship feature coefficient, and the relationship coefficient between the user and the recipient/sender.
  • the importance of the text received or sent by the user is recognized according to the content of the text, the relationship between the user and the recipient and sender of the text, and the closeness of the relationship between the user and the recipient and/or the sender of the text.
  • the determining module is specifically configured to use the content feature coefficient and the relationship feature coefficient as input data of a machine learning method, and determine the importance of the text through the machine learning method. Based on the above technical solution, after the content feature coefficient and the relationship feature coefficient are obtained, the content feature coefficient and the relationship feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method. Identify the importance of text received or sent by the user.
  • the determining module is specifically configured to use the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient as the input data of the machine learning method, and the importance of the text is determined by the machine learning method .
  • the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient can be used as the input data of the machine learning method , Determine the importance of the text through machine learning methods, so that the importance of the text received or sent by the user can be recognized.
  • the determination module is specifically configured to use the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the user-recipient/sender relationship coefficient as the input data of the machine learning method , Through the machine learning method to determine the importance of the text.
  • the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender are used as input data of the machine learning method, and the importance of the text is determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the device further includes: a sorting module; the processing module is further used for the coefficient in the content feature coefficient The sum and the coefficient in the relationship feature coefficient are multiplied to obtain the importance coefficient of the text; the sorting module is used to sort the text received or sent by the user according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to obtain the importance of the text. It also sorts the text received or sent by the user according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient includes N coefficients
  • the device further includes: a sorting module; the processing module is further used for the coefficient in the content feature coefficient The sum is multiplied by the coefficient in the relationship feature coefficient, and the message is divided by the feature coefficient of the influence range of the message to obtain the importance coefficient of the text; the sorting module is used to receive or receive the user according to the importance coefficient of the text The sent text is sorted.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to affect the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, the content feature coefficient includes N coefficients, and the device further includes: a sorting module; the processing module is further used for the coefficient in the content feature coefficient The sum is multiplied by the coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender, and divided by the message influence range feature coefficient to obtain the importance coefficient of the text; the ranking module uses In order to sort the text received or sent by the user according to the importance coefficient of the text.
  • the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the user and the recipient can be combined.
  • /Sender’s relationship coefficient is multiplied, and the message’s influence range feature coefficient is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text, so that the user can be based on the text The importance of processing text.
  • the determining module is specifically used to multiply the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; the determining module also specifically uses To determine the importance of the text according to the importance coefficient of the text. Based on the above technical solution, after the content feature coefficient and the relation feature coefficient are obtained, the sum of the coefficients in the content feature coefficient and the coefficient in the relation feature coefficient can be multiplied to obtain the importance coefficient of the text.
  • the importance coefficient of the text determines the importance of the text so that the importance of the text received or sent by the user can be recognized.
  • the determining module is specifically used to multiply the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients, and divide with the feature coefficients of the message influence range to obtain the text Importance coefficient; the determining module is also specifically used to determine the importance of the text according to the importance coefficient of the text.
  • the determination module is specifically configured to multiply the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the relationship coefficients between the user and the receiver/sender, and The feature coefficient of the influence range of the message is divided to obtain the importance coefficient of the text; the determining module is also specifically used to determine the importance of the text according to the importance coefficient of the text.
  • the sum of the content characteristic coefficients can be combined with The coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender are multiplied, and the message influence range feature coefficient is divided by the coefficient of the text to obtain the importance coefficient of the text, and the text is determined according to the importance coefficient of the text.
  • the importance of the text which can identify the importance of the text received or sent by the user.
  • the device when there are multiple texts with the same importance, the device further includes: a sorting module; the sorting module is used to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the sorting module is used to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the device further includes: a display module; the display module is used to classify and display the text according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • the present application provides an apparatus for recognizing text.
  • the apparatus may include: at least one processor and a memory, the memory storing a software program, and the processor is configured to call the software program in the memory to execute the following process: Obtain text and portrait information of the user from the memory or server, where the text is the text received or sent by the user, and the portrait information of the user is used to indicate N keywords in the text related to the user, where N is greater than Or an integer equal to 1, the server is connected to the device for recognizing text, and the text and portrait information of the user are stored in the server; the content feature coefficient is obtained according to the text and the portrait information of the user, wherein the content feature coefficient Used to indicate the importance of the content of the text; determine the importance of the text according to the content feature coefficient.
  • the technical solution provided by the seventh aspect above can obtain text and user portrait information, obtain content feature coefficients based on the text and user portrait information, and determine the importance of the text according to the content feature coefficients, thereby identifying users based on the content of the text The importance of the text received or sent.
  • the portrait information of the user includes N tag information, and the N tag information correspondingly indicate the N keywords; or, the portrait information of the user includes the N tag information and the N tags
  • the weight of each tag information in the message can be obtained according to the text and N tag information, and the importance of the text can be determined according to the content feature coefficients, or it can be based on the text, N tag information, and the value of each tag information in the N tag information.
  • the weight obtains the content feature coefficient, and determines the importance of the text according to the content feature coefficient, so that the importance of the text received or sent by the user can be recognized according to the content of the text.
  • the content feature coefficient includes N coefficients, where the nth coefficient is obtained according to the nth label information in the text and the portrait information of the user, and n is a positive value greater than 0 and less than or equal to N. Integer.
  • the nth coefficient in the content feature coefficients can be obtained according to the nth label information in the text and user portrait information, and the importance of the text can be determined according to the content feature coefficients, so that the user can identify whether the user received or received according to the content of the text. Importance of the text sent.
  • the obtaining content feature coefficients according to the text and the portrait information of the user includes: obtaining first user behavior feedback data from the memory or the server, where the first user behavior feedback data is used to indicate The importance of the text with a high degree of similarity to the text content in history; obtain the first correspondence from the memory or the server, where the first correspondence is used to indicate the first user behavior feedback data and the first user The relationship between the influence coefficients of the behavior feedback data; the influence coefficient of the first user behavior feedback data is obtained according to the first user behavior feedback data and the first correspondence; according to the text, the portrait information of the user, and the first correspondence An influence coefficient of user behavior feedback data obtains the content characteristic coefficient.
  • the first user behavior feedback data can be obtained, the influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data and the first relationship, and the influence coefficient of the first user behavior feedback data can be obtained according to the text, the portrait information of the user, and the first user behavior
  • the influence coefficient of the feedback data obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient, so that the importance of the text received or sent by the user can be recognized according to the content of the text.
  • the processor is further configured to call a software program in the memory to execute the following process: obtain user relationship information from the memory or the server, where the user relationship information is used to indicate that the user is connected with The hierarchical relationship of other users; the relationship feature coefficient is obtained according to the text and the relationship information of the user, where the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text; the determination The importance of the text also includes: determining the importance of the text according to the relationship feature coefficient.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user's subordinate relationship information, the user's department relationship information, the collaboration relationship information, and the user's communication relationship information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the frequency of communication between the user and other users, and the content feature coefficient and relationship
  • the feature coefficient determines the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the relationship feature coefficient can be determined according to the content feature coefficient and the relationship feature coefficient
  • the importance of the text so that the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the subordinate relationship feature coefficient according to the user's subordinate relationship information obtains the department relationship feature coefficient according to the user's department relationship information, obtain the synergy relationship feature coefficient according to the collaboration relationship information, and obtain the communication based on the user's communication relationship information Relationship feature coefficient, and determine the importance of the text based on the content feature coefficient, the subordinate relationship feature coefficient, the department relationship feature coefficient, the collaboration relationship feature coefficient, and the communication relationship feature coefficient, so that the text content, the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the relationship characteristic coefficients include organizational relationship characteristic coefficients and communication relationship characteristic coefficients.
  • the organizational relationship characteristic coefficients include subordinate relationship characteristic coefficients, department relationship characteristic coefficients, and synergy relationship characteristic coefficients; according to the text and the Obtaining the relationship feature coefficient from the relationship information of the user includes: obtaining second user behavior feedback data from the memory or the server, where the second user behavior feedback data is used to indicate the importance of the text that is the same as the sender of the text in history.
  • the second correspondence is used to indicate the second user behavior feedback data and the correlation between the influence coefficient of the second user behavior feedback data; according to the second user behavior feedback
  • the data and the second corresponding relationship obtain the influence coefficient of the second user behavior feedback data; the relationship characteristic coefficient is obtained according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence. According to the text, the user's relationship information, and the first 2.
  • the influence coefficient of the user behavior feedback data obtains the relation characteristic coefficient, and the importance of the text is determined according to the content characteristic coefficient and the relation characteristic coefficient, so that the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text Or the importance of the text sent.
  • the text includes the identification of the recipient of the text; the processor is further configured to call the software program in the memory to execute the following process: obtain the message influence range characteristic coefficient according to the identification of the recipient of the text , Wherein the message influence range characteristic coefficient is used to indicate the number of recipients of the text; determining the importance of the text further includes: determining the importance of the text according to the message influence range characteristic coefficient.
  • the message influence range characteristic coefficient can be obtained according to the text, and the importance of the text can be determined according to the content characteristic coefficient, the relationship characteristic coefficient and the message influence range characteristic coefficient, so that the content of the text, the user and the recipient of the text can be determined
  • the relationship with the sender and the number of recipients of the text identify the importance of the text received or sent by the user.
  • the processor is further configured to call a software program in the memory to execute the following process: obtain the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, where the user and the receiver
  • the relationship coefficient of the sender is used to indicate the closeness of the relationship between the user and the recipient of the text and/or the sender of the text; determining the importance of the text also includes: according to the user and the recipient/sender
  • the human relationship coefficient determines the importance of the text.
  • the relationship coefficient between the user and the receiver/sender can be obtained according to the relationship information of the user, and the importance of the text can be determined according to the content feature coefficient, the relationship feature coefficient, and the relationship coefficient between the user and the receiver/sender.
  • the importance of the text received or sent by the user is recognized according to the content of the text, the relationship between the user and the recipient and sender of the text, and the closeness of the relationship between the user and the recipient and/or the sender of the text.
  • a possible implementation manner is that determining the importance of the text according to the content feature coefficient includes: using the content feature coefficient as input data of a machine learning method, and determining the importance of the text through the machine learning method. Based on the above technical solution, after the content feature coefficient is obtained, the content feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized .
  • the determining the importance of the text according to the content feature coefficients includes: using the content feature coefficients and the relationship feature coefficients as input data of a machine learning method, and determining the importance of the text through the machine learning method . Based on the above technical solution, after the content feature coefficients and relationship feature coefficients are obtained, the content feature coefficients and relationship feature coefficients can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the user can recognize that the user receives or Importance of the text sent.
  • the determining the importance of the text according to the content feature coefficient includes: using the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient as the input data of the machine learning method, and the machine learning Method to determine the importance of the text.
  • the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient can be used as the input data of the machine learning method , Determine the importance of the text through machine learning methods, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficients, including: using the content feature coefficients, the relationship feature coefficients, the message influence range feature coefficients, and the relationship coefficients between the user and the recipient/sender As the input data of the machine learning method, the importance of the text is determined by the machine learning method.
  • the content feature coefficients, relationship feature coefficients, message influence range feature coefficients, and user-recipient/sender relationship coefficients is used as the input data of the machine learning method, and the importance of the text is determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the content feature coefficient when there are multiple texts with the same importance, includes N coefficients, and the processor is further configured to call the software program in the memory to execute the following process: All or part of the N coefficients in are added to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • all or part of the N coefficients in the content feature coefficients can be added to obtain the importance coefficient of the text, and according to the text
  • the importance coefficient ranks the text received or sent by the user, which enables the user to process the text according to the importance of the text.
  • the content feature coefficient when there are multiple texts with the same importance, the content feature coefficient includes N coefficients, and the processor is further configured to call the software program in the memory to execute the following process: The sum of the coefficients in and the coefficients in the relationship feature coefficient are multiplied to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the processor is further configured to call the software program in the memory to execute the following process: The sum of the coefficients in and the coefficients in the relationship feature coefficient are multiplied to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the content feature coefficient includes N coefficients
  • the processor is further configured to call the software program in the memory to execute the following process: Multiply the sum of the coefficients in with the coefficients in the relationship feature coefficient, and divide with the message’s influence range feature coefficient to get the importance coefficient of the text; the text received or sent by the user according to the importance coefficient of the text put in order.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to affect the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient includes N coefficients
  • the processor is further configured to call the software program in the memory to execute the following process: Multiply the sum of the coefficients in the relationship feature coefficients and the relationship coefficients between the user and the recipient/sender, and divide it with the message influence range feature coefficient to get the importance coefficient of the text; according to the The importance coefficient of the text ranks the text received or sent by the user.
  • the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the user and recipient/ The relationship coefficient of the sender is multiplied, and the feature coefficient of the influence range of the message is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text. Importance of processing text.
  • a possible implementation manner is to determine the importance of the text according to the content feature coefficients, including: adding all or part of the N coefficients in the content feature coefficients to obtain the importance coefficient of the text; according to the importance of the text
  • the gender factor determines the importance of the text.
  • the determining the importance of the text according to the content feature coefficients includes: multiplying the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text ; Determine the importance of the text according to the importance coefficient of the text. Based on the above technical solution, after the content feature coefficient and the relationship feature coefficient are obtained, the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied to obtain the importance coefficient of the text.
  • the importance coefficient of determines the importance of the text, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficients, including: multiplying the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients, and the message influence range feature coefficients Do the division operation to get the importance coefficient of the text; determine the importance of the text according to the importance coefficient of the text.
  • the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied by The feature coefficient of the message influence range is divided to obtain the importance coefficient of the text, and the importance of the text is determined according to the importance coefficient of the text, so that the importance of the text received or sent by the user can be recognized.
  • determining the importance of the text according to the content feature coefficients includes: the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the relationship between the user and the recipient/sender The relationship coefficient is multiplied and divided by the characteristic coefficient of the influence range of the message to obtain the importance coefficient of the text; the importance of the text is determined according to the importance coefficient of the text.
  • the sum of the content characteristic coefficients can be combined with The coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender are multiplied, and the message influence range feature coefficient is divided by the coefficient of the text to obtain the importance coefficient of the text, and the text is determined according to the importance coefficient of the text.
  • the importance of the text which can identify the importance of the text received or sent by the user.
  • the processor when there are multiple texts with the same importance, the processor is also used to call the software program in the memory to execute the following process: according to the importance coefficient of the text, the user receives or sends The text is sorted. Based on the above technical solution, after determining the importance of the text, when there are multiple texts with the same importance, the text received or sent by the user can be sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text .
  • the processor is further configured to call a software program in the memory to execute the following process: the text is classified and displayed according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • the present application provides an apparatus for recognizing text.
  • the apparatus may include: at least one processor and a memory, the memory stores a software program, and the processor invokes the software program in the memory to execute the following process:
  • the memory or server obtains text and user relationship information, where the text is the text received or sent by the user, the user relationship information is used to indicate the hierarchical relationship between the user and other users, and the relationship between the server and the recognized text
  • the device is connected, the server stores the text and the relationship information of the user; the relationship feature coefficient is obtained according to the text and the relationship information of the user, wherein the relationship feature coefficient is used to indicate the user and the recipient of the text and/ Or the relationship of the sender of the text; determine the importance of the text according to the relationship feature coefficient.
  • the technical solution provided by the above eighth aspect can obtain the relationship information of the text and the user, obtain the relationship feature coefficient according to the relationship information of the text and the user, and determine the importance of the text according to the relationship feature coefficient, so that the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user’s superior and subordinate relationship information, the user’s department relationship information, the collaboration relationship information, and the user’s communication relationship information, and the importance of the text can be determined according to the relationship feature coefficient.
  • the relationship between the user and the recipient and sender of the text identifies the importance of the text received or sent by the user.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the communication frequency between the user and other users, and the text can be determined according to the relationship feature coefficient Therefore, the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the importance of the text can be determined according to the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the relationship characteristic coefficients include organizational relationship characteristic coefficients and communication relationship characteristic coefficients.
  • the organizational relationship characteristic coefficients include subordinate relationship characteristic coefficients, department relationship characteristic coefficients, and synergy relationship characteristic coefficients; according to the text and the Obtaining the relationship feature coefficient from the relationship information of the user includes: obtaining second user behavior feedback data from the memory or the server, where the second user behavior feedback data is used to indicate the importance of the text that is the same as the sender of the text in history. ⁇ ; acquiring a second correspondence, where the second correspondence is used to indicate the association between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; according to the second user behavior feedback data And the second corresponding relationship, the influence coefficient of the second user behavior feedback data is obtained; the relationship characteristic coefficient is obtained according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence.
  • the text the user's relationship information, and the first 2.
  • the influence coefficient of the user behavior feedback data obtains the relationship feature coefficient, and the importance of the text is determined according to the relationship feature coefficient, so that the importance of the text received or sent by the user can be identified according to the relationship between the user and the recipient and sender of the text.
  • a possible implementation manner is that determining the importance of the text according to the relationship feature coefficient includes: using the relationship feature coefficient as input data of a machine learning method, and determining the importance of the text through the machine learning method. Based on the above technical solution, after the relationship feature coefficient is obtained, the relationship feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so that the importance of the text received or sent by the user can be recognized.
  • the processor when there are multiple texts with the same importance, is also used to call the software program in the memory to execute the following process: multiply the coefficients in the relation characteristic coefficients to obtain the text The importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the coefficient in the relation feature coefficient can be multiplied to obtain the importance coefficient of the text, and the user can be Sorting the received or sent text allows users to process the text according to the importance of the text.
  • determining the importance of the text according to the relationship feature coefficient includes: multiplying the coefficients in the relationship feature coefficient to obtain the importance coefficient of the text; and determining the importance coefficient of the text according to the importance coefficient of the text. The importance of text. Based on the above technical solutions, after the relationship feature coefficients are obtained, the coefficients in the relationship feature coefficients can be multiplied to obtain the importance coefficient of the text, and the importance of the text can be determined according to the importance coefficient of the text, so as to identify the user receiving or sending The importance of the text.
  • the processor when there are multiple texts with the same importance, the processor is also used to call the software program in the memory to execute the following process: according to the importance coefficient of the text, the user receives or sends The text is sorted. Based on the above technical solution, after determining the importance of the text, when there are multiple texts with the same importance, the text received or sent by the user can be sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text .
  • the processor is further configured to call a software program in the memory to execute the following process: the text is classified and displayed according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • the present application provides an apparatus for recognizing text.
  • the apparatus may include: at least one processor and a memory, the memory storing a software program, and the processor is configured to call the software program in the memory to execute the following process: Obtain text, user portrait information, and user relationship information from the memory or server, where the text is the text received or sent by the user, and the user portrait information is used to indicate the N keys in the user-related text A word, N is an integer greater than or equal to 1.
  • the user's relationship information is used to indicate the hierarchical relationship between the user and other users.
  • the server is connected to the text recognition device.
  • the server stores text, user portrait information, and The relationship information of the user; the content feature coefficient is obtained according to the text and the portrait information of the user, wherein the content feature coefficient is used to indicate the importance of the content of the text; the relationship feature coefficient is obtained according to the text and the relationship information of the user, where The relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text; the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient.
  • the technical solution provided by the above-mentioned ninth aspect can obtain text, user portrait information, and user relationship information, obtain content feature coefficients based on text and user portrait information, obtain relationship feature coefficients based on text and user relationship information, and obtain relationship feature coefficients based on content
  • the feature coefficient and the relationship feature coefficient determine the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the portrait information of the user includes N tag information, and the N tag information correspondingly indicate the N keywords; or, the portrait information of the user includes the N tag information and the N tags
  • the weight of each tag information in the message can be obtained according to the text and N tag information, and the importance of the text can be determined according to the content feature coefficients and the relationship feature coefficients, or the text, N tag information, and each of the N tag information
  • the weight of each tag information obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient, so that the content received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text The importance of text.
  • the content feature coefficient includes N coefficients, where the nth coefficient is obtained according to the nth label information in the text and the portrait information of the user, and n is a positive value greater than 0 and less than or equal to N. Integer.
  • the nth coefficient in the content feature coefficient can be obtained according to the nth label information in the text and user portrait information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient, so that the content of the text can be determined , And the relationship between the user and the recipient and sender of the text to identify the importance of the text received or sent by the user.
  • Obtaining the content feature coefficients according to the text and the portrait information of the user includes: obtaining first user behavior feedback data from the memory or the server, where the first user behavior feedback data is used to indicate history The importance of the text with high similarity to the text content; obtain the first correspondence from the memory or the server, where the first correspondence is used to indicate the first user behavior feedback data and the first user behavior feedback The correlation between the influence coefficients of the data; the influence coefficient of the first user behavior feedback data is obtained according to the first user behavior feedback data and the first correspondence; according to the text, the portrait information of the user, and the first user The influence coefficient of the behavior feedback data obtains the content characteristic coefficient.
  • the first user behavior feedback data can be obtained, the influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data and the first relationship, and the influence coefficient of the first user behavior feedback data can be obtained according to the text, the portrait information of the user, and the first user behavior
  • the influence coefficient of the feedback data obtains the content feature coefficient, and the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient, so that the content received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text The importance of text.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration Relationship information
  • the user’s subordinate relationship information is used to indicate the subordinate relationship between the user and other users
  • the user’s department relationship information is used to indicate the minimum unit organization of the user and other users, and the minimum unit organization is the user and In the same organization where other users are located, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works collaboratively with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the user's subordinate relationship information, the user's department relationship information, the collaboration relationship information, and the user's communication relationship information, and the importance of the text can be determined according to the content feature coefficient and the relationship feature coefficient In this way, the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the user's superior-subordinate relationship with other users;
  • the user's department relationship information includes the minimum unit organization of the user and other users;
  • the collaboration relationship information includes instruction information,
  • the indication information is used to indicate whether the user and other users work together;
  • the communication relationship information of the user includes the frequency of communication between the user and other users.
  • the relationship feature coefficient can be obtained according to the text, the subordinate relationship between the user and other users, the smallest unit organization between the user and other users, the instruction information, and the frequency of communication between the user and other users, and the content feature coefficient and relationship
  • the feature coefficient determines the importance of the text, so that the importance of the text received or sent by the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text.
  • the user’s superior-subordinate relationship information includes the weight of the superior-subordinate relationship between the user and other users, and the weight of the superior-subordinate relationship between the user and other users is used to indicate the superior-subordinate relationship between the user and other users.
  • Level relationship the user’s department relationship information includes the weight of the smallest unit organization, which is used to indicate the smallest unit organization of the user and other users;
  • the collaboration relationship information includes the weight of the collaboration relationship, and the weight of the collaboration relationship is used for Indicates the collaboration relationship between the user and other users;
  • the communication relationship information of the user includes the weight of the communication frequency, and the weight of the communication frequency is used to indicate the frequency of communication with other users.
  • the relationship feature coefficient can be obtained according to the text, the weight of the subordinate relationship between the user and other users, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency, and the relationship feature coefficient can be determined according to the content feature coefficient and the relationship feature coefficient
  • the importance of the text so that the importance of the text received or sent by the user can be identified based on the content of the text and the relationship between the user and the recipient and sender of the text.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the synergy relationship feature coefficient is obtained based on the synergy relationship information
  • the communication relationship feature coefficient is based on the The user’s communication relationship information is obtained.
  • the subordinate relationship feature coefficient according to the user's subordinate relationship information obtains the department relationship feature coefficient according to the user's department relationship information, obtain the synergy relationship feature coefficient according to the collaboration relationship information, and obtain the communication based on the user's communication relationship information Relationship feature coefficient, and determine the importance of the text based on the content feature coefficient, the subordinate relationship feature coefficient, the department relationship feature coefficient, the collaboration relationship feature coefficient, and the communication relationship feature coefficient, so that the text content, the user and the recipient of the text can be determined
  • the relationship with the sender identifies the importance of the text received or sent by the user.
  • the relationship characteristic coefficients include organizational relationship characteristic coefficients and communication relationship characteristic coefficients.
  • the organizational relationship characteristic coefficients include subordinate relationship characteristic coefficients, department relationship characteristic coefficients, and synergy relationship characteristic coefficients; according to the text and the Obtaining the relationship feature coefficient from the relationship information of the user includes: obtaining second user behavior feedback data from the memory or the server, where the second user behavior feedback data is used to indicate the importance of the text that is the same as the sender of the text in history.
  • the second correspondence is used to indicate the correlation between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; according to The second user behavior feedback data and the second corresponding relationship obtain the influence coefficient of the second user behavior feedback data; the relationship feature is obtained according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data coefficient.
  • the second user behavior feedback data can be obtained, and the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data and the second correspondence. According to the text, the user's relationship information, and the first 2.
  • the influence coefficient of the user behavior feedback data obtains the relation characteristic coefficient, and the importance of the text is determined according to the content characteristic coefficient and the relation characteristic coefficient, so that the user can be identified according to the content of the text and the relationship between the user and the recipient and sender of the text Or the importance of the text sent.
  • the processor is further configured to call a software program in the memory to execute the following process: obtain the message influence range characteristic coefficient according to the text, wherein the message influence range characteristic coefficient is used to indicate the content of the text The number of recipients; the determining the importance of the text also includes: determining the importance of the text according to the characteristic coefficient of the influence range of the message.
  • the message influence range characteristic coefficient can be obtained according to the text, and the importance of the text can be determined according to the content characteristic coefficient, the relationship characteristic coefficient and the message influence range characteristic coefficient, so that the content of the text, the user and the recipient of the text can be determined
  • the relationship of the sender and the number of recipients of the text identify the importance of the text received or sent by the user.
  • the processor is further configured to call a software program in the memory to execute the following process: obtain the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, where the user and the receiver The person/sender relationship coefficient is used to indicate the closeness of the relationship between the user and the recipient of the text and/or the sender of the text; determining the importance of the text also includes: according to the user and the recipient/ The sender's relationship coefficient determines the importance of the text.
  • the relationship coefficient between the user and the recipient/sender can be obtained according to the relationship information of the user, and the importance of the text can be determined according to the content feature coefficient, the relationship feature coefficient, and the relationship coefficient between the user and the recipient/sender.
  • the importance of the text received or sent by the user is recognized according to the content of the text, the relationship between the user and the recipient and sender of the text, and the closeness of the relationship between the user and the recipient and/or the sender of the text.
  • the determining the importance of the text according to the content feature coefficient and the relation feature coefficient includes: using the content feature coefficient and the relation feature coefficient as input data of a machine learning method, and determining the text by the machine learning method The importance of text. Based on the above technical solution, after the content feature coefficient and the relationship feature coefficient are obtained, the content feature coefficient and the relationship feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method. Identify the importance of text received or sent by the user.
  • the determining the importance of the text according to the content feature coefficient and the relationship feature coefficient includes: using the content feature coefficient, the relationship feature coefficient and the message influence range feature coefficient as the input data of the machine learning method, Determine the importance of the text through the machine learning method.
  • the content feature coefficient, the relationship feature coefficient and the message influence range feature coefficient can be used as the input data of the machine learning method, and the importance of the text can be determined by the machine learning method, so as to identify The importance of the text received or sent by the user.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficient and the relationship feature coefficient, including: using the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the user and the recipient/sender
  • the human relationship coefficient is used as the input data of the machine learning method, and the importance of the text is determined by the machine learning method.
  • the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender can be used as the input data of the machine learning method.
  • the learning method determines the importance of the text, so that the importance of the text received or sent by the user can be recognized.
  • the content feature coefficient when there are multiple texts with the same importance, the content feature coefficient includes N coefficients, and the processor is further configured to call the software program in the memory to execute the following process: The sum of the coefficients in and the coefficients in the relationship feature coefficient are multiplied to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the processor is further configured to call the software program in the memory to execute the following process: The sum of the coefficients in and the coefficients in the relationship feature coefficient are multiplied to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • the content feature coefficient includes N coefficients
  • the processor is further configured to call the software program in the memory to execute the following process: Multiply the sum of the coefficients in with the coefficients in the relationship feature coefficient, and divide with the message’s influence range feature coefficient to get the importance coefficient of the text; the text received or sent by the user according to the importance coefficient of the text put in order.
  • the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients can be multiplied to affect the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text.
  • the content feature coefficient includes N coefficients
  • the processor is further configured to call the software program in the memory to execute the following process: Multiply the sum of the coefficients in the relationship feature coefficients and the relationship coefficients between the user and the recipient/sender, and divide it with the message influence range feature coefficient to get the importance coefficient of the text; according to the The importance coefficient of the text ranks the text received or sent by the user.
  • the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the user and recipient/ The relationship coefficient of the sender is multiplied, and the feature coefficient of the influence range of the message is divided to obtain the importance coefficient of the text, and the text received or sent by the user is sorted according to the importance coefficient of the text. Importance of processing text.
  • a possible implementation manner is to determine the importance of the text according to content feature coefficients and relation feature coefficients, including: multiplying the sum of the coefficients in the content feature coefficients and the coefficients in the relation feature coefficients to obtain the text Importance coefficient; the importance of the text is determined according to the importance coefficient of the text.
  • the content feature coefficient and the relationship feature coefficient are obtained, the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied to obtain the importance coefficient of the text.
  • the importance coefficient of determines the importance of the text, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation manner is to determine the importance of the text according to the content feature coefficients and the relation feature coefficients, including: multiplying the sum of the coefficients in the content feature coefficients with the coefficients in the relation feature coefficients, and the message
  • the influence range feature coefficient is divided to obtain the importance coefficient of the text; the importance of the text is determined according to the importance coefficient of the text.
  • the sum of the content feature coefficients and the coefficients in the relationship feature coefficient can be multiplied by
  • the feature coefficient of the message influence range is divided to obtain the importance coefficient of the text, and the importance of the text is determined according to the importance coefficient of the text, so that the importance of the text received or sent by the user can be recognized.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficient and the relation feature coefficient, including: the sum of the coefficients in the content feature coefficient, the coefficient in the relation feature coefficient, and the user and the recipient
  • the relationship coefficient of the sender is multiplied and divided by the characteristic coefficient of the influence range of the message to obtain the importance coefficient of the text; the importance of the text is determined according to the importance coefficient of the text.
  • the sum of the content characteristic coefficients can be combined with The coefficient in the relationship feature coefficient and the relationship coefficient between the user and the recipient/sender are multiplied, and the message influence range feature coefficient is divided by the coefficient of the text to obtain the importance coefficient of the text, and the text is determined according to the importance coefficient of the text.
  • the importance of the text which can identify the importance of the text received or sent by the user.
  • the processor when there are multiple texts with the same importance, the processor is also used to call the software program in the memory to execute the following process: according to the importance coefficient of the text, the user receives or sends The text is sorted. Based on the above technical solution, after determining the importance of the text, when there are multiple texts with the same importance, the text received or sent by the user can be sorted according to the importance coefficient of the text, so that the user can process the text according to the importance of the text .
  • the processor is further configured to call a software program in the memory to execute the following process: the text is classified and displayed according to the importance of the text. Based on the above technical solution, the text can be classified and displayed according to the importance of the text, so that the user can process the text of different importance types separately.
  • the present application provides a system chip that can be used in an apparatus for recognizing text.
  • the system chip includes: at least one processor, and related program instructions are executed in the at least one processor to implement The text recognition method described in the first aspect and any one of the possible implementation manners of the first aspect.
  • the system chip may further include at least one memory, and the memory stores related program instructions.
  • this application provides a system chip that can be used in a text recognition device.
  • the system chip includes: at least one processor, and related program instructions are executed in the at least one processor to The method for recognizing text as described in any two possible implementations of the first aspect and the second aspect is implemented.
  • the system chip may further include at least one memory, and the memory stores related program instructions.
  • the present application provides a system chip that can be used in an apparatus for recognizing text.
  • the system chip includes: at least one processor, and related program instructions are executed in the at least one processor to The method for recognizing the text described in the third aspect and any one of the possible implementation manners of the third aspect is implemented.
  • the system chip may further include at least one memory, and the memory stores related program instructions.
  • this application provides a computer storage medium that can be used in a text recognition device.
  • the computer-readable storage medium stores program instructions. When the program instructions are run, The text recognition method described in the first aspect and various possible implementations of the first aspect.
  • this application provides a computer storage medium that can be used in a text recognition device.
  • the computer-readable storage medium stores program instructions. When the program instructions are run, The method for recognizing text described in the second aspect and various possible implementations of the second aspect.
  • this application provides a computer storage medium that can be used in a text recognition device.
  • the computer-readable storage medium stores program instructions. When the program instructions are run, The method for text recognition described in the third aspect and various possible implementation manners of the third aspect.
  • the present application provides a computer program product that contains program instructions, and when the program instructions involved are executed, the program instructions described in the first aspect and various possible implementation manners of the first aspect Method of recognizing text.
  • the present application provides a computer program product that contains program instructions, and when the program instructions involved are executed, the program instructions described in the second aspect and various possible implementation manners of the second aspect Method of recognizing text.
  • the present application provides a computer program product, which contains program instructions, and when the program instructions involved are executed, the program instructions described in the third aspect and various possible implementation manners of the third aspect Method of recognizing text.
  • any system chip, computer storage medium, or computer program product provided above is used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the corresponding method in The beneficial effects will not be repeated here.
  • Figure 1 is a schematic diagram of a display interface provided by an embodiment of the application.
  • FIG. 2 is a schematic diagram of the hardware structure of a hardware device provided by an embodiment of the application.
  • FIG. 3 is a first schematic flowchart of a method for recognizing text provided by an embodiment of this application;
  • FIG. 4 is a second schematic flowchart of a method for recognizing text provided by an embodiment of this application.
  • FIG. 5 is a third schematic flowchart of a method for recognizing text provided by an embodiment of this application.
  • FIG. 6 is a fourth flowchart of a method for recognizing text provided by an embodiment of this application.
  • FIG. 7 is a schematic flowchart five of the method for recognizing text provided by an embodiment of this application.
  • FIG. 8 is a first structural diagram of a text recognition device provided by an embodiment of this application.
  • FIG. 9 is a second structural diagram of a text recognition device provided by an embodiment of the application.
  • FIG. 10 is a third structural diagram of a text recognition device provided by an embodiment of this application.
  • FIG. 11 is a fourth structural diagram of a text recognition device provided by an embodiment of the application.
  • FIG. 12 is a schematic diagram 5 of the structure of a text recognition device provided by an embodiment of the application.
  • FIG. 13 is a sixth structural diagram of a text recognition device provided by an embodiment of this application.
  • FIG. 14 is a seventh structural diagram of a text recognition device provided by an embodiment of this application.
  • FIG. 15 is a schematic diagram eight of the structure of a text recognition device provided by an embodiment of this application.
  • FIG. 16 is a schematic diagram 9 of the structure of a text recognition apparatus provided by an embodiment of the application.
  • mobile office software for example, Tencent, eSpace, Outlook and other software
  • Tencent, eSpace, Outlook and other software has become an indispensable application in people's daily work.
  • users receive or send a large amount of text every day, but among these large amounts of text, only part of the text is user-related. Therefore, users need to filter out user-related text from a large amount of text, and how to efficiently filter out user-related text from a large amount of text is an urgent problem to be solved.
  • an embodiment of the present application provides a method for recognizing texts.
  • the method can be applied to a device installed with mobile office software. After the text is written, the text and the portrait information of the user can be obtained, the content feature coefficients can be obtained according to the text and the user portrait information, and the importance of the text can be determined according to the content feature coefficients. In this way, the importance of the text can be determined based on the content of the text. Therefore, it is possible to automatically filter user-related texts from a large amount of texts, which is more efficient than manual filtering of user-related texts in the prior art.
  • the identifier of the recipient of the text and/or the identifier of the sender of the text can be obtained, as well as the user's relationship information, according to the identification of the recipient of the text and/or the identification of the sender of the text , And the user's relationship information to obtain the relationship feature coefficient, and determine the importance of the text according to the relationship feature coefficient.
  • the importance of the text can be determined according to the relationship between the user and the recipient of the text and/or the sender of the text. Therefore, it is possible to automatically filter user-related texts from a large amount of texts, which is more efficient than manual filtering of user-related texts in the prior art.
  • FIG. 1 is a schematic diagram of a display interface provided by an embodiment of this application.
  • the user has received text 1-text N, a total of N texts.
  • the device for recognizing text can separately obtain the portrait information of each text and the user, obtain the content feature coefficient of each text according to the portrait information of each text and the user, and determine the importance of each text according to the content feature coefficient of each text.
  • text 2, text 6, text 7, and text 5 are important texts (the text marked with an exclamation mark in Figure 1), and the rest are non-important texts.
  • the text may be the text received or sent by the user, and the text may include the identifier of the recipient, the identifier of the sender, and the content of the text.
  • the text may be an instant message received by the user through RTX, or an email sent by the user through the mailbox.
  • the following embodiments of the present application are introduced by taking the text as the text received by the user as an example. If the text is the text sent by the user, reference may be made to the introduction of the text received by the user in the following embodiments of the present application, and will not be repeated.
  • the portrait information of the user can be used to indicate the N keywords in the text related to the user, where N is an integer greater than or equal to 1.
  • the portrait information of the user can be used to indicate the occurrence in the text related to the user in history. N words with higher frequency.
  • the user's portrait information can be used to indicate the names of important items.
  • the portrait information of the user may include N pieces of label information, and the N pieces of label information correspondingly indicate N keywords.
  • ⁇ tag 1 , tag 2 ,..., tag N ⁇ or the portrait information of the user may include N pieces of tag information and the weight of each tag information in the N pieces of tag information, for example, ⁇ tag 1 , tag 2 ,... , Tag N ⁇ and ⁇ tagVal 1 , tagVal 2 ,..., tagVal N ⁇ , where tagVal 1 is the weight of tag 1 , tagVal 2 is the weight of tag 2 ...tagVal N is the weight of tag N.
  • the label information may include one or more characters.
  • the portrait information of the user can be set by the user or obtained based on the text sent or received in the user's history.
  • obtain important texts in the user's history extract words with higher frequency from the important texts, use words with higher frequency as tag information, and set the weight of each tag information according to the frequency of each tag information , The higher the frequency of the label information, the greater the weight of the label information.
  • the content feature coefficient can be used to indicate the importance of the content of the text.
  • the content feature coefficient may include N coefficients, and the nth coefficient may be obtained according to the nth label information in the text and the user's portrait information.
  • n is an integer greater than 0 and less than or equal to N.
  • the identifier of the recipient may be the user name of the recipient or the ID of the recipient.
  • the ID of the recipient may include numbers and letters.
  • the identifier of the sender can be the user name of the sender or the ID of the sender.
  • the sender's ID can include numbers and letters.
  • the user's relationship information may include the user's organizational relationship information and the user's communication relationship information.
  • the organization relationship information of the user may be used to indicate the hierarchical relationship between the user and other users in the organization (hereinafter abbreviated as user and other users).
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration relationship information.
  • the user's superior-subordinate relationship information can be used to indicate the user's superior-subordinate relationship with other users.
  • the superior-subordinate relationship between the user and other users may include the user's superior, the user's peer, the user's subordinate, and there is no superior-subordinate relationship, and so on.
  • the user's superior-subordinate relationship information includes the user's superior-subordinate relationship with other users.
  • user 1’s superior-subordinate relationship information may include ⁇ user 1, user 2, superior ⁇ ; ⁇ user 1, user 3, same level ⁇ ; ⁇ user 1, user 4, subordinate ⁇ ; ⁇ user 1, user 5, no There is a superior-subordinate relationship ⁇ and so on, where user 2 is the superior user of user 1, user 3 is the same-level user of user 1, user 4 is a subordinate user of user 1, and user 5 is a user who does not have a superior-subordinate relationship with user 1. .
  • the user's superior-subordinate relationship information includes the weight of the user's superior-subordinate relationship with other users.
  • the weight of the subordinate relationship can be used to represent the subordinate relationship between the user and other users.
  • the weight of the subordinate relationship between a user and other users may include: the weight of the subordinate relationship between the user and the superior user, the weight of the subordinate relationship between the user and the user at the same level, and the subordinate relationship between the user and the subordinate user.
  • the weight of the subordinate relationship between the user and the superior user is greater than the weight of the subordinate relationship between the user and the user at the same level
  • the subordinate relationship between the user and the same user is greater than the weight of the subordinate relationship between the user and the subordinate user.
  • the weight of the subordinate relationship of the subordinate user is greater than the weight of the subordinate relationship between the user and the user who does not have the subordinate relationship.
  • the weight of the subordinate relationship between the user and the superior user is X
  • the weight of the subordinate relationship between the user and the user at the same level is Y
  • the weight of the superior relationship between the user and the subordinate user is Z
  • the subordinate relationship of user 1 The relationship information includes ⁇ User 1, User 2, X ⁇ ; ⁇ User 1, User 3, Y ⁇ ; ⁇ User 1, User 4, Z ⁇ ; ⁇ User 1, User 5, W ⁇ , etc.
  • User 2 Is the upper level of user 1
  • X is the weight of the upper-lower relationship between user 1 and user 2
  • user 3 is the same level of user as user 1
  • Y is the weight of the upper-lower relationship between user 1 and user 3
  • user 4 is user 1
  • Z is the weight of the subordinate relationship between user 1 and user 4
  • user 5 is a user who does not have a subordinate relationship with user 1
  • W is the weight of the subordinate relationship between user 1 and user 5, where X>Y >Z>W.
  • X, Y, Z, and W are positive numbers.
  • the subordinate relationship between a user and other users can also include other subordinate relationships, for example, the user’s superior superior and the user
  • the embodiments of this application are not limited to the subordinates of the subordinates.
  • the user's superior-subordinate relationship information may also be in other forms, such as a list form, which is not limited in the embodiment of the present application.
  • the user's departmental relationship information can be used to indicate the minimum unit organization between the user and other users.
  • the smallest unit organization of users and other users can be described as the organization with the least number of users in the same organization where the user and other users are located.
  • the types of the smallest unit organization may include: small organization, large organization, and first organization, etc.
  • the large organization may include multiple small organizations
  • the first organization may include multiple large organizations.
  • the smallest unit organization of user 1 is small organization 1 (that is, the organization where user 1 is located, the organization with the least number of users is small organization 1)
  • the smallest unit organization of user 2 is small organization 1
  • the smallest unit organization of user 3 Is small organization 2
  • small organization 1 and small organization 2 are groups in large organization 1
  • the smallest unit organization of user 1 and user 2 is small organization 1
  • the smallest unit organization of user 1 and user 3 is large organization 1.
  • the smallest unit organization of 2 and user 3 is also large organization 1.
  • the department relationship information of the user includes the minimum unit organization of the user and other users.
  • the department relationship information of user 1 may include ⁇ user 1, user 2, small organization 1 ⁇ ; ⁇ user 1, user 3, small organization 1 ⁇ ; ⁇ user 1, user 4, large organization 1 ⁇ , etc., among which,
  • the smallest unit organization of user 1 and user 2 is small organization 1
  • the smallest unit organization of user 1 and user 3 is also small organization 1
  • the smallest unit organization of user 1 and user 4 is large organization 1.
  • the department relationship information of the user includes the weight of the smallest unit organization.
  • the weight of the smallest unit organization may be the weight of the smallest unit organization of the user and other users.
  • the weight of the minimum unit organization can be used to indicate the minimum unit organization of the user and other users.
  • the weight of the smallest unit organization may include the weight of the small organization, the weight of the large organization, and the weight of the first organization. Among them, the weight of the small organization is more important than the weight of the large organization, and the weight of the large organization is more important than the weight of the first organization.
  • the weight of the small organization is A
  • the weight of the large organization is B
  • the weight of the first organization is C.
  • the department relationship information of user 1 includes ⁇ user 1, user 2, A ⁇ ; ⁇ user 1, user 3 , B ⁇ ; ⁇ User 1, User 4, C ⁇ etc.
  • User 2 is a user of the small organization where User 1 is located
  • A is the weight of the smallest unit organization of User 1 and User 2
  • User 3 is not User 1 User of the small organization where User 3 is a user of the large organization where User 1 is located
  • B is the weight of the smallest unit organization of User 1 and User 3
  • User 4 is not a user of the small or large organization where User 1 is located, but , User 4 is a user of the first organization where User 1 is located
  • C is the weight of the smallest unit organization of User 1 and User 4, where A>B>C.
  • A, B, and C are positive numbers.
  • the user's department relationship information may also be in other forms, such as a list form, which is not limited in the embodiment of the present application.
  • the type of the smallest unit organization may also include other types, which are not limited in the embodiment of the present application.
  • the collaboration relationship information can be used to indicate whether the user works in collaboration with other users.
  • the collaboration relationship information includes indication information, where the indication information is used to indicate whether the user works in coordination with other users.
  • the indication information may be 1-bit indication information. For example, when the indication information is 0, it can indicate that the user is not working with other users, and when the indication information is 1, it can indicate that the user is working with other users, and vice versa.
  • the collaboration relationship information of user 1 may include ⁇ user 1, user 2, 0 ⁇ ; ⁇ user 1, user 3, 1 ⁇ ; ⁇ user 1, user 4, 1 ⁇ , etc., among which, the relationship between user 1 and user 2 There is no collaborative work, user 1 and user 3 are working together, and user 1 and user 4 are working together.
  • the collaboration relationship information includes a collaboration relationship weight
  • the collaboration relationship weight may be a collaboration relationship weight between the user and other users.
  • the weight of the collaboration relationship between the user and other users may indicate the collaboration relationship between the user and other users.
  • the synergy relationship weight may include: the synergy relationship weight when the user has a synergy relationship with other users, and the synergy relationship weight when the user has no synergy relationship with other users.
  • the weight of the collaborative relationship when the user has a collaborative relationship with other users is greater than the weight of the collaborative relationship when the user has no collaborative relationship with other users.
  • the collaborative relationship information of user 1 may include ⁇ user 1, user 2, P ⁇ ; ⁇ User 1, User 3, M ⁇ ; ⁇ User 1, User 4, M ⁇ , etc., among which, user 1 and user 2 work together, user 1 and user 3 do not work together, user 1 and user 4 did not work together.
  • P and M are positive numbers.
  • the weight of the collaborative relationship may include: the weight of the collaborative relationship when the user has no collaborative relationship with other users, the weight of the collaborative relationship when the user has a collaborative relationship with other users on important items, and the weight of the collaborative relationship between the user and other users.
  • the weight of the collaborative relationship when there is a collaborative relationship on the project is greater than the weight of the collaborative relationship when the user has a collaborative relationship with other users on non-important items, and the user has a collaborative relationship with other users on non-important items
  • the weight of the collaborative relationship when the user has no collaborative relationship with other users is H
  • the weight of the collaborative relationship when the user has a collaborative relationship with other users on important items is I
  • the user has collaboration with other users on non-important items For example, the collaborative relationship weight of the relationship is J.
  • the collaborative relationship information of user 1 can include ⁇ User 1, User 2, H ⁇ ; ⁇ User 1, User 3, I ⁇ ; ⁇ User 1, User 4, J ⁇ , etc. , Among them, user 1 and user 2 are not working together, user 1 and user 3 are working together on important projects, and user 1 and user 4 are working together on non-important projects.
  • I, J, and H are positive numbers.
  • the user's communication relationship information can be used to indicate the frequency of communication between the user and other users.
  • the user's communication relationship information includes a communication frequency, where the communication frequency is a communication frequency between the user and other users. For example, if the user 1 and the user 2 send or receive 20 pieces of information in one day, the communication relationship information of the user 1 is 20.
  • the user's communication relationship information includes the weight of communication frequency.
  • the weight of the communication frequency is the weight of the communication frequency between the user and other users.
  • the weight of the communication frequency between the user and other users is used to indicate the communication frequency between the user and other users.
  • the weight of the communication frequency is positively related to the frequency of communication between the user and other users, that is, the higher the frequency of communication between the user and other users, the greater the weight of the communication frequency, the lower the frequency of communication between the user and other users, and the lower the communication frequency. The smaller the weight.
  • the weight of the communication frequency is R. If the frequency of communication between the user and other users is greater than or equal to 11 times/day , And less than or equal to 30 times/day, the weight of the communication frequency is S. If the user’s communication frequency with other users is greater than or equal to 31 times/day, the weight of the communication frequency is T, where T>S>R, T, S and R is a positive number.
  • the relationship feature coefficient can be used to indicate the relationship between the user and the recipient and/or the sender.
  • the relationship characteristic coefficient may include an organization relationship characteristic coefficient and a communication relationship characteristic coefficient.
  • the organizational relationship feature coefficients may include subordinate relationship feature coefficients, department relationship feature coefficients, and collaboration relationship feature coefficients.
  • the subordinate relationship feature coefficients can include the subordinate relationship feature coefficients between the user and each recipient.
  • the subordinate relationship feature coefficients between the user and each recipient can be used to indicate the relationship between the user and each recipient.
  • the department relationship characteristic coefficient may include the department relationship characteristic coefficient between the user and each recipient, and the department relationship characteristic coefficient between the user and each recipient may be used to indicate the minimum unit organization of the user and each recipient.
  • the collaborative relationship feature coefficient may include the collaborative relationship feature coefficient between the user and each recipient, and the collaborative relationship feature coefficient between the user and each recipient may be used to indicate whether the user and each recipient work in cooperation.
  • the characteristic coefficient of the communication relationship may include the characteristic coefficient of the communication relationship between the user and each recipient, and the characteristic coefficient of the communication relationship between the user and each recipient may be used to indicate the frequency of communication between the user and each recipient.
  • the upper-lower relationship feature coefficients can include the user-sender's superior-subordinate relationship feature coefficients, and the user-sender's superior-subordinate relationship feature coefficients can be used to indicate the superior-subordinate relationship between the user and the sender relationship.
  • the department relationship characteristic coefficient may include the department relationship characteristic coefficient of the user and the sender, and the department relationship characteristic coefficient of the user and the sender may be used to indicate the minimum unit organization of the user and the sender.
  • the collaborative relationship feature coefficient may include the collaborative relationship feature coefficient between the user and the sender, and the collaborative relationship feature coefficient between the user and the sender can be used to indicate whether the user and the sender are working together.
  • the characteristic coefficient of the communication relationship may include the characteristic coefficient of the communication relationship between the user and the sender, and the characteristic coefficient of the communication relationship between the user and the sender may be used to indicate the frequency of communication between the user and the sender.
  • the subordinate relationship feature coefficients may include the subordinate relationship feature coefficients between the user and the sender, and the subordinate relationship feature coefficients between the user and recipients other than the user.
  • the characteristic coefficients of the subordinate relationship of recipients other than the user may be used to indicate the subordinate relationship between the user and the recipients other than the user.
  • the department relationship feature coefficient may include the department relationship feature coefficient between the user and the sender, and the department relationship feature coefficient between the user and the recipient other than the user.
  • the department relationship feature coefficient between the user and the recipient other than the user can be used for Indicates the smallest unit organization of the user and the recipient other than the user.
  • the collaborative relationship feature coefficients may include the collaborative relationship feature coefficients between the user and the sender, and the collaborative relationship feature coefficients between the user and recipients other than the user.
  • the collaborative relationship feature coefficients between the user and the recipient other than the user can be used to indicate Whether the user and recipients other than the user work together.
  • the communication relationship feature coefficients can include the communication relationship feature coefficients between the user and the sender, and the communication relationship feature coefficients between the user and the recipient other than the user.
  • the communication relationship feature coefficient between the user and the recipient other than the user can be used to indicate The frequency of communication between users and recipients other than users.
  • the apparatus for recognizing text provided in the embodiment of the present application may be a functional module in a device. It is understandable that the above functions can be components in hardware devices, such as chips in mobile phones, software functions running on dedicated hardware, or virtualization functions instantiated on platforms (for example, cloud platforms) .
  • the apparatus for recognizing text may be implemented by the hardware device 200 in FIG. 2.
  • Fig. 2 shows a schematic diagram of the hardware structure of a hardware device applicable to the embodiments of the present application.
  • the hardware device 200 includes at least one processor 201, a communication line 202, a memory 203, and at least one communication interface 204.
  • the processor 201 can be a general-purpose central processing unit (central processing unit, CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more programs for controlling the execution of the program of this application. integrated circuit.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • the communication line 202 may include a path for transferring information between the aforementioned components, such as a bus.
  • the communication interface 204 uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet interfaces, radio access network interfaces (RAN), and wireless local area networks (wireless local area networks, WLAN) etc.
  • a transceiver uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet interfaces, radio access network interfaces (RAN), and wireless local area networks (wireless local area networks, WLAN) etc.
  • the memory 203 can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM), or other types that can store information and instructions
  • the dynamic storage device can also be electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disk storage, optical disc storage (Including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or store desired program codes in the form of instructions or data structures and can be used by a computer Any other media accessed, but not limited to this.
  • the memory can exist independently and is connected to the processor through the communication line 202.
  • the memory can also be integrated with the processor.
  • the memory provided in the embodiments of the present application may generally be non-volatile.
  • the memory 203 is used to store and execute the computer execution instructions involved in the solution of the present application, and the processor 201 controls the execution.
  • the processor 201 is configured to execute computer-executable instructions stored in the memory 203, so as to implement the method provided in the embodiment of the present application.
  • the computer-executable instructions in the embodiments of the present application may also be referred to as application program codes, which are not specifically limited in the embodiments of the present application.
  • the processor 201 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 2.
  • the hardware device 200 may include multiple processors, such as the processor 201 and the processor 207 in FIG. 2. Each of these processors can be a single-CPU (single-CPU) processor or a multi-core (multi-CPU) processor.
  • the processor here may refer to one or more devices, circuits, and/or processing cores for processing data (for example, computer program instructions).
  • the hardware device 200 may further include an output device 205 and an input device 206.
  • the output device 205 communicates with the processor 201 and can display information in a variety of ways.
  • the output device 205 may be a liquid crystal display (LCD), a light emitting diode (LED) display device, a cathode ray tube (CRT) display device, or a projector (projector) Wait.
  • the input device 206 communicates with the processor 201 and can receive user input in a variety of ways.
  • the input device 206 may be a mouse, a keyboard, a touch screen device, a sensor device, or the like.
  • the aforementioned hardware device 200 may be a general-purpose device or a special-purpose device.
  • the hardware device 200 may be a desktop computer, a portable computer, a PDA (personal digital assistant, PDA), a mobile phone, a tablet computer, a wireless terminal device, an embedded device, or a device with a similar structure in FIG. 2.
  • PDA personal digital assistant
  • the embodiment of the present application does not limit the type of the hardware device 200.
  • the device for recognizing text can execute part or all of the steps in the embodiment of this application. These steps are only examples, and the embodiment of this application can also execute other steps or variations of various steps. In addition, each step may be executed in a different order presented in the embodiment of the present application, and it may not be necessary to perform all the steps in the embodiment of the present application.
  • the device for recognizing the text can obtain the text and the portrait information of the user, obtain the content feature coefficient according to the text and the user portrait information, and determine the importance of the text according to the content feature coefficient, so that the text content can be determined according to the content of the text. importance.
  • the method for recognizing text may include steps 301-303.
  • Step 301 Obtain text and user portrait information.
  • obtaining the text and the portrait information of the user includes: obtaining the text and the portrait information of the user from a remote site (for example, a server) and/or locally.
  • a remote site for example, a server
  • the server can be connected to a text recognition device, and the server stores text and/or portrait information of the user.
  • the processor 201 calls the interface to obtain the text and the portrait information of the user from the memory 203.
  • the processor 201 calls the interface to obtain the text from the memory 203, and the processor 201 sends a request to the service through the communication interface 204, and the request includes the user's identification.
  • the server sends the portrait information of the user, and the processor 201 receives the portrait information of the user through the communication interface 204.
  • the content of the text can be used to replace the text in step 301. Subsequently, the content feature coefficient can be obtained according to the content of the text and the portrait information of the user.
  • Step 302 Obtain content feature coefficients according to the text and the user's portrait information.
  • the description of the content feature coefficient can refer to the above-mentioned noun explanation, which will not be repeated here.
  • the content feature coefficient is positively correlated with the importance of the text, that is, when other feature coefficients remain unchanged, the larger the content feature coefficient, the higher the importance of the text.
  • step 302 can refer to the following example:
  • Example 1 If the user's portrait information includes N pieces of label information, the content feature coefficient can be determined according to the number of times the label information in the user's portrait information appears in the text.
  • the coefficient corresponding to the nth label information can be 1; if the nth label information in the user’s portrait information does not appear in the text, the The coefficient corresponding to n pieces of label information may be zero.
  • the content feature coefficients can be determined according to the correspondence shown in Table 2.
  • Table 2 if the nth label information in the user's portrait information does not appear in the text, the coefficient corresponding to the nth label information can be 0; if the nth label information in the user's portrait information appears in the text If the number of times is greater than or equal to 1 time and less than or equal to 8, the coefficient corresponding to the nth label information can be 1; if the number of times the nth label information in the user's portrait information appears in the text is greater than or equal to 9 times, then the nth The coefficient corresponding to the label information can be 2.
  • Example 2 If the user’s portrait information includes N pieces of label information and the weight of each piece of label information in the N pieces of label information, the content can be determined according to the number of times the label information appears in the text in the user’s portrait information and the weight of the label information Characteristic coefficient.
  • the first parameter of the nth label information can be obtained according to the number of times the nth label information in the user’s portrait information appears in the text, and the product of the first parameter and the weight of the nth label information can be taken as the nth The coefficient corresponding to the label information.
  • the user’s portrait information includes ⁇ tag 1 , tag 2 , tag 3 ⁇ and ⁇ tagVal 1 , tagVal 2 , tagVal 3 ⁇ , tag 1 and tag 2 do not appear in the text, tag 3 appears in the text, If the tag information appears in the text, the first parameter of the tag information is 1. If the tag information does not appear in the text, the first parameter of the tag information is 0, for example, the first parameters of tag 1 and tag 2 If the value is 0, the first parameter of tag 3 is 1, the coefficient corresponding to tag 1 and tag 2 is 0, and the coefficient corresponding to tag 3 is tagVal 3. Therefore, the content feature coefficient is ⁇ 0, 0, tagVal 3 ⁇ .
  • Example 3 If the user's portrait information includes N tag information, the text keywords can be extracted from the text, the similarity between the text keywords and each tag information can be obtained, and the similarity can be used as the content feature coefficient.
  • the keywords of the text may include words and/or phrases extracted from the text.
  • word segmentation software for example, jieba software, ansj software, or HanLP software, etc.
  • the keywords of the text can include one or more words and/or phrases.
  • the keywords of the text include tag information or synonyms (or synonyms) of tag information.
  • the similarity between the keywords of the text and the label information can be used to indicate the degree of similarity between the keywords of the text and the label information.
  • the similarity between the keywords of the text and the label information can include the same and different, and can also include the same, different, and similar.
  • the keywords of the text can be compared with the label information. If the characters included in the two are exactly the same, it is determined that the keywords of the text and the label information are the same; if the number of characters that are not the same in the characters included in the two is greater than or Equal to the first preset threshold, it is determined that the keywords of the text are not the same as the label information; if there are different characters in the characters included in the two, and the number of different characters is less than or equal to the second preset threshold, it is determined The keywords of the text are similar to the label information.
  • 0, 1, and 2 may be used to indicate the similarity between the keywords of the text and the tag information.
  • the similarity between the keywords of the text and the label information is 1, and the coefficient corresponding to the label information is 1.
  • the keywords of the text are not the same as the label information, the The similarity between the keyword and the tag information is 0, and the coefficient corresponding to the tag information is 0.
  • the similarity between the keywords of the text and the label information can also be expressed in other ways or characters, which is not limited.
  • Example 4 If the user’s portrait information includes N tag information and the weight of each tag information in the N tag information, the keywords of the text can be extracted from the text to obtain the similarity between the keywords of the text and each tag information.
  • the nth similarity degree is multiplied with the weight of the label information corresponding to the similarity degree to obtain the nth second parameter, and the nth second parameter is used as the nth coefficient in the content feature coefficient.
  • the portrait information of the user includes ⁇ tag 1 , tag 2 , ..., tag N ⁇ and ⁇ tagVal 1 , tagVal 2 , ..., tagVal N ⁇ , the keywords of the text and each tag information in the user's portrait information
  • the similarity is ⁇ similarCoff 1 , similarCoff 2 ,..., similarCoff N ⁇ (similarCoff 1 is the similarity between the text keyword and tag 1 , and similarCoff 2 is the similarity between the text keyword and tag 2 , and so on) is
  • the n-th second parameter is used as the parameter of the first function
  • the value of the first function is used as the n-th coefficient in the content characteristic coefficient, wherein the derivative of the function is greater than and equal to zero.
  • the portrait information of the user includes ⁇ tag 1 , tag 2 , ..., tag N ⁇ and ⁇ tagVal 1 , tagVal 2 , ..., tagVal N ⁇ , the keywords of the text and each tag information in the user's portrait information
  • the similarity is ⁇ similarCoff 1 , similarCoff 2 ,..., similarCoff N ⁇ (similarCoff 1 is the similarity between the text keyword and tag 1 , and similarCoff 2 is the similarity between the text keyword and tag 2 , and so on) is
  • the content feature coefficient can be obtained not only based on the text and the portrait information of the user, but also based on the text, the portrait information of the user, and the first user behavior feedback data.
  • the first user behavior feedback data can be used to indicate the importance of a text with high similarity to the text content in history.
  • the first user behavior feedback data can be obtained in step 301 or before step 302.
  • the influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data.
  • the influence coefficient of the first user behavior feedback data can be used The content feature coefficient is corrected to make the corrected content feature coefficient more accurate. Specifically, you can refer to the description in Example 5 below.
  • Example 5 The influence coefficient of the first user behavior feedback data can be obtained according to the first user behavior feedback data, and the content feature coefficient obtained by the method described in Example 1 to Example 4 above and the influence coefficient of the first user behavior feedback data can be added or subtracted Calculate to get more accurate content feature coefficients.
  • the influence coefficient of the first user behavior feedback data may be used to indicate the importance of the text with high similarity to the text content in history.
  • the influence coefficient of the first user behavior feedback data is positively related to the importance of the text, that is, under the condition that other feature coefficients remain unchanged, the greater the influence coefficient of the first user behavior feedback data, the higher the importance of the text .
  • obtaining the influence coefficient of the first user behavior feedback data according to the first user behavior feedback data includes: acquiring a first correspondence, where the first correspondence is used to indicate that the influence coefficient of the first user behavior feedback data is The association relationship between the first user behavior feedback data; the influence coefficient of the first user behavior feedback data is determined according to the first user behavior feedback data and the first corresponding relationship.
  • the first correspondence relationship may be preset and stored in the memory 203 or the server in FIG. 2.
  • acquiring the first correspondence relationship includes acquiring the first correspondence relationship from the memory 203 or the server in FIG. 2.
  • the content feature coefficients obtained by the method described in Example 1 to Example 4 above are ⁇ contentCoff 1 , contentCoff 2 ,..., contentCoff N ⁇ , the influence coefficient of the first user behavior feedback data and the first user behavior feedback data
  • the corresponding relationship is shown in Table 3 as an example. If the first user behavior feedback data indicates that the historically similar text with the text is an important text, the influence coefficient of the first user behavior feedback data is a, and the first user behavior feedback data is considered
  • a new content feature coefficient after the influence coefficient of the user behavior feedback data is ⁇ contentCoff 1 +a, contentCoff 2 +a,..., contentCoff N +a ⁇ .
  • Step 303 Determine the importance of the text according to the content feature coefficient.
  • One possible implementation manner is to determine the importance of the text according to the content feature coefficients, including: using the content feature coefficients as input data of the machine learning method, and determining the importance of the text through the machine learning method.
  • the importance of text can take many forms, for example, important text and non-important text.
  • Another example is the first-level important text, the second-level important text, the third-level important text, and the non-important text.
  • the first-level important text is more important than the second-level important text
  • the second-level important text is more important than the third-level text.
  • the importance of important texts, the importance of three-level important texts is greater than the importance of non-important texts.
  • machine learning methods can include traditional machine learning methods and deep learning methods.
  • Traditional machine learning methods may include: linear discriminant analysis (LDA), support vector machine (support vector machine, SVM), enhanced learning method (adboost), and so on.
  • LDA linear discriminant analysis
  • SVM support vector machine
  • Adboost enhanced learning method
  • the deep learning methods may include: deep neural networks (DNN), convolutional neural networks (convolutional neural networks, CNN), recurrent neural networks (recurrent neural networks, RNN), and so on.
  • DNN deep neural networks
  • CNN convolutional neural networks
  • RNN recurrent neural networks
  • the text when there are multiple texts with the same importance (for example, when there are multiple first-level important texts), after determining the importance of the text according to the content feature coefficients, it also includes: adding all the N coefficients in the content feature coefficients Or part of the addition is performed to obtain the importance coefficient of the text; the text received or sent by the user is sorted according to the importance coefficient of the text.
  • textImportantCoff is the importance coefficient of the text
  • contentCoff 1 , contentCoff 2 , ..., contentCoff N are content feature coefficients.
  • Another possible implementation method is to determine the importance of the text according to the content feature coefficients, including: adding all or part of the N coefficients in the content feature coefficients to obtain the importance coefficient of the text; determining the text according to the importance coefficient of the text The importance of.
  • determining the importance of the text according to the importance coefficient of the text includes: if the importance coefficient of the text is greater than or equal to the first threshold, determining that the text is an important text; or, according to the importance coefficient of the text and the The corresponding relationship of importance determines the importance of the text.
  • the importance coefficient of the text As shown in Table 4, it is the correspondence between the importance coefficient of the text and the importance of the text.
  • the importance coefficient of the text is greater than or equal to 9, the importance of the text is the first-level important text. If the importance coefficient of the text is 6-8, the importance of the text is the second-level important text. If the importance coefficient of is 2-5, the importance of the text is a three-level important text. If the importance coefficient of the text is 0-1, the importance of the text is non-important text.
  • Table 4 is only an example of the correspondence between the importance coefficient of the text and the importance of the text.
  • the correspondence between the importance coefficient of the text and the importance of the text can also be in other forms, and
  • the corresponding relationship between the importance coefficient of the text and the importance of the text can be a certain line in Table 1, some lines, all of Table 1, or more corresponding relationships than those shown in Table 1. This application will not proceed. Specific restrictions.
  • the method further includes: sorting the text received or sent by the user according to the importance coefficient of the text.
  • the text is classified and displayed according to the importance of the text.
  • the processor 201 in FIG. 2 can sort and display the text in the output device 205 according to the importance of the text by calling an interface (for example, display important text in front of non-important text).
  • the importance of the text in the old device can be transferred to the new device through the following two methods:
  • Method 1 Using display instructions, extract the text identification, text importance and/or text importance coefficient from the data storage area of the old device, and encrypt, and then encrypt the encrypted text identification, text importance and/ Or the importance coefficient of the text is transmitted to the server through a secure network transmission method, and the new device can download the identification of the encrypted text, the importance of the text and/or the importance coefficient of the text from the server through a display instruction.
  • Method 2 Copy the text identification, the importance of the text, and/or the importance coefficient of the text in the old device to the new device by means of direct copy.
  • the text and user portrait information can be obtained, the content feature coefficients can be obtained according to the text and user portrait information, and the importance of the text can be determined according to the content feature coefficient, so that the importance of the text can be determined according to the content of the text Sex.
  • the user After the user receives the text, he can obtain the recipient’s identity and/or sender’s identity, as well as the user’s relationship information, and obtain the relationship feature coefficients according to the recipient’s identity and/or sender’s identity and the user’s relationship information.
  • the relationship feature coefficient determines the importance of the text, so that the importance of the text can be determined according to the relationship between the user and the recipient and/or the sender.
  • the method for recognizing text may include steps 401-403.
  • Step 401 Obtain the identifier of the recipient and/or the identifier of the sender, and the relationship information of the user.
  • the identifier of the receiver and/or the identifier of the sender, and the description of the user's relationship information can refer to the introduction of the above terms, which will not be repeated here.
  • obtaining the recipient’s identity and/or the sender’s identity and the user’s relationship information includes: obtaining the recipient’s identity and/or the sender’s identity from a remote site (for example, a server) and/or locally , And the user's relationship information.
  • a remote site for example, a server
  • the server can be connected to the device for recognizing text, and the server stores the identifier of the recipient and/or the sender, and/or the relationship information of the user.
  • the following describes obtaining the identifier of the recipient and/or the identifier of the sender, and the relationship information of the user in conjunction with FIG. 2.
  • the processor 201 calls the interface to obtain the identifier of the recipient and/or the sender, and the relationship information of the user from the memory 203.
  • the processor 201 calls the interface to obtain the recipient’s identity and/or the sender’s identity from the memory 203, and the processor 201 sends the The service sending request includes the user's identification. After receiving the request, the server sends the user's relationship information, and the processor 201 receives the user's relationship information through the communication interface 204.
  • the recipient if the user is the sender, obtain the recipient’s identity and user relationship information; if the user is the only recipient of the text, obtain the sender’s identity and user relationship information; if the user is among multiple recipients One is to obtain the recipient's identity, sender's identity, and user relationship information.
  • the user's relationship information when there is no personnel change, the user's relationship information is generally fixed. Therefore, the user's relationship information can be obtained when the text is sent or received for the first time, and subsequently, when receiving or sending a new text
  • the user's relationship information can be obtained when the text is sent or received for the first time, and subsequently, when receiving or sending a new text
  • the administrator can update the change to the local or server, and send a notification message to the text-recognizing device, and the text-recognizing device receives the text Get the user's relationship information again after the notification message.
  • Step 402 Obtain the relationship feature coefficient according to the identifier of the receiver and/or the identifier of the sender, and the relationship information of the user.
  • the relationship feature coefficient is positively related to the importance of the text, that is, when other feature coefficients remain unchanged, the larger the relationship feature coefficient, the higher the importance of the text.
  • the user's relationship information may include the user's organizational relationship information and the user's communication relationship information.
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration relationship information.
  • the user's communication relationship information can refer to the introduction of the above terms, and will not be repeated.
  • Relationship characteristic coefficients include: organizational relationship characteristic coefficients and communication relationship characteristic coefficients.
  • the organizational relationship feature coefficients may include subordinate relationship feature coefficients, department relationship feature coefficients, and collaboration relationship feature coefficients.
  • the description of the subordinate relationship feature coefficient, department relationship feature coefficient, synergy relationship feature coefficient, and communication relationship feature coefficient can refer to the introduction of the above terms, and will not be repeated.
  • the organizational relationship feature coefficient can be obtained according to the user's organizational relationship information.
  • the communication relationship characteristic coefficient can be obtained according to the user's communication relationship information.
  • the organizational relationship feature coefficient can be obtained based on the user's organizational relationship information, including: obtaining the subordinate relationship feature coefficient based on the user's subordinate relationship information, obtaining the department relationship feature coefficient based on the user's department relationship information, and obtaining it based on the collaboration relationship information Characteristic coefficient of synergy.
  • the characteristic coefficients of the superior-subordinate relationship can be obtained in the following four situations:
  • Case 1 The user is the only recipient, and the user's superior-subordinate relationship information includes the user's superior-subordinate relationship with other users.
  • the upper-lower relationship characteristic coefficients may include the upper-lower relationship characteristic coefficients between the user and the sender. According to the user’s superior-subordinate relationship information, obtain the superior-subordinate relationship characteristic coefficients, including: obtaining the superior-subordinate relationship between the user and the sender from the superior-subordinate relationship between the user and other users, and obtain the user and sender based on the superior-subordinate relationship between the user and the sender The characteristic coefficient of a person's superior-subordinate relationship.
  • a possible implementation method is to obtain the characteristic coefficient of the upper-lower relationship between the user and the sender according to the upper-lower relationship between the user and the sender, including: according to the upper-lower relationship between the user and the sender, according to the setting rules of the upper-lower relationship characteristic coefficient Determine the characteristic coefficient of the subordinate relationship between the user and the sender.
  • the setting rules for the upper-lower relationship feature coefficients can be as follows: the upper-lower relationship feature coefficients between the user and the upper-level user are greater than the upper-lower relationship feature coefficients between the user and the same-level user, and the upper-lower relationship feature coefficients between the user and the same-level user.
  • the characteristic coefficient of the upper-lower relationship between the user and the lower-level user, and the characteristic coefficient of the upper-lower relationship between the user and the lower-level user is greater than the characteristic coefficient of the upper-lower relationship between the user and the user without the upper-lower relationship.
  • the subordinate relationship information of user 1 includes ⁇ user 1, user 2, superior ⁇ ; ⁇ user 1, user 3, same level ⁇ ; ⁇ user 1, user 4, subordinate ⁇ ; ⁇ user 1, user 5 ,
  • the feature coefficient of the superior-subordinate relationship between the user and the user's superior user is 6
  • the feature coefficient of the superior-subordinate relationship between the user and the user's same-level user is 4,
  • the feature coefficient of the superior-subordinate relationship between the user and the user's subordinate user The coefficient is 2.
  • the feature coefficient of the superior-subordinate relationship between the user and the user is 0.01, for example, if the sender's identity is user 2, the superior-subordinate relationship feature coefficient between the user and the sender can be 6; if If the sender’s identity is user 3, the characteristic coefficient of the subordinate relationship between the user and the sender can be 4; if the sender’s identity is user 4, the characteristic coefficient of the subordinate relationship between the user and the sender can be 2; if the sender is The identifier of is user 5, and the characteristic coefficient of the superior-subordinate relationship between the user and the sender can be 0.01.
  • Case 2 The user is the only recipient, and the user's superior-subordinate relationship information may include the weight of the user's superior-subordinate relationship with other users.
  • the upper-lower relationship characteristic coefficients may include the upper-lower relationship characteristic coefficients between the user and the sender.
  • Obtain the characteristic coefficients of the superior-subordinate relationship according to the user's superior-subordinate relationship information including: obtaining the weight of the superior-subordinate relationship between the user and the sender from the weight of the superior-subordinate relationship between the user and other users, and according to the subordinate relationship between the user and the sender The weight is used to obtain the characteristic coefficient of the upper-lower relationship between the user and the sender.
  • a possible implementation method is to obtain the characteristic coefficients of the superior-subordinate relationship between the user and the sender according to the weight of the subordinate relationship between the user and the sender, including: taking the subordinate relationship between the user and the sender as the weight of the user and the sender The characteristic coefficient of the subordinate relationship.
  • the subordinate relationship information of user 1 includes ⁇ user 1, user 2, X ⁇ ; ⁇ user 1, user 3, Y ⁇ ; ⁇ user 1, user 4, Z ⁇ ; ⁇ user 1, user 5, W ⁇ , where X is the weight of the subordinate relationship between user 1 and user 2, Y is the weight of the subordinate relationship between user 1 and user 3, Z is the weight of the subordinate relationship between user 1 and user 4, and W is the user Take the weight of the upper-lower relationship between 1 and user 5 as an example.
  • the characteristic coefficient of the upper-lower relationship between the user and the sender can be X; if the sender’s identity is user 3, the user and sender The characteristic coefficient of the upper-lower relationship of a person can be Y; if the sender’s identity is user 4, the characteristic coefficient of the upper-lower relationship between the user and the sender can be Z; if the identity of the sender is user 5, the The upper-lower relationship feature coefficient can be W.
  • the weight of the subordinate relationship between the user and the sender can also be processed (for example, adding One number or minus one number, etc.), and use the processed data as the characteristic coefficient of the upper-lower relationship between the user and the sender.
  • the processed data should meet the above-mentioned setting rules for the relationship feature coefficients of the upper and lower levels.
  • Case 3 The user is one of multiple recipients, and the user's superior-subordinate relationship information may include the user's superior-subordinate relationship with other users.
  • the subordinate relationship feature coefficients include the subordinate relationship feature coefficients of the user and the sender, and the subordinate relationship feature coefficients of the user and recipients other than the user.
  • the specific description of the characteristic coefficients of the superior-subordinate relationship between the user and the sender can be referred to the introduction in the above case 1, and will not be repeated.
  • the specific description of the characteristic coefficient of the upper-lower relationship between the user and the recipient other than the user can refer to the above case 1.
  • the upper-lower relationship, the introduction of the characteristic coefficients of the upper-lower relationship between the user and the sender will not be repeated.
  • Case 4 The user is one of multiple recipients, and the user's superior-subordinate relationship information may include the weight of the user's superior-subordinate relationship with other users.
  • the subordinate relationship feature coefficients include the subordinate relationship feature coefficients of the user and the sender, and the subordinate relationship feature coefficients of the user and recipients other than the user.
  • the specific description of the characteristic coefficients of the superior-subordinate relationship between the user and the sender can be referred to the introduction in the above case 2, and will not be repeated.
  • the specific description of the characteristic coefficient of the upper-lower relationship between the user and the recipient other than the user can refer to the above case 2.
  • the weight of the person's superior-subordinate relationship is introduced by the characteristic coefficients of the superior-subordinate relationship between the user and the sender, and will not be repeated.
  • the department relationship characteristic coefficient can be obtained in the following 4 situations:
  • Case 5 The user is the only recipient, and the user's departmental relationship information includes the smallest unit organization between the user and other users.
  • the department relationship characteristic coefficient includes the department relationship characteristic coefficient between the user and the sender.
  • a possible implementation method is to obtain the department relationship characteristic coefficient between the user and the sender according to the minimum unit organization of the user and the sender, including: determining the user according to the department relationship characteristic coefficient setting rule according to the minimum unit organization of the user and the sender The characteristic coefficient of the department relationship with the sender.
  • the department relationship feature coefficient setting rules can be as follows: the department relationship feature coefficient of users in a small organization is greater than that of users in a large organization, and the department relationship feature coefficient of users in a large organization is greater than that of the first organization The characteristic coefficient of the department relationship of the user in.
  • the department relationship information of user 1 includes ⁇ user 1, user 2, small organization 1 ⁇ ; ⁇ user 1, user 3, small organization 1 ⁇ ; ⁇ user 1, user 4, large organization 1 ⁇ , small organization
  • the department relationship feature coefficient of users in the middle is 5, the department relationship characteristic coefficient of users in a large organization is 2, and the department relationship characteristic coefficient of users in the first organization is 0.01 as an example. If the sender’s identity is user 2, then The department relationship feature coefficient between the user and the sender can be 5; if the sender’s identity is user 3, the department relationship characteristic coefficient between the user and the sender can be 2; if the sender’s identity is user 4, the user and the sender The departmental relationship feature coefficient of can be 0.01.
  • Case 6 The user is the only recipient, and the department relationship information of the user includes the weight of the smallest unit organization between the user and other users.
  • the department relationship characteristic coefficient includes the department relationship characteristic coefficient between the user and the sender.
  • Obtain department relationship feature coefficients according to the user’s department relationship information including: obtaining the weight of the user’s and sender’s smallest unit organization from the weight of the user’s and other users’ smallest unit organization, and according to the weight of the user’s and sender’s smallest unit organization, Obtain the characteristic coefficient of the department relationship between the user and the sender.
  • a possible implementation method is to obtain the department relationship characteristic coefficients between the user and the sender according to the weight of the minimum unit organization of the user and the sender, including: taking the weight of the minimum unit organization of the user and the sender as the department of the user and the sender Relationship characteristic coefficient.
  • the department relationship information of user 1 includes ⁇ user 1, user 2, A ⁇ ; ⁇ user 1, user 3, B ⁇ ; ⁇ user 1, user 4, C ⁇ , where A is user 1 and user 2 is the weight of the smallest unit organization, B is the weight of the smallest unit organization of user 1 and user 3, C is the weight of the smallest unit organization of user 1 and user 4, for example, if the sender’s identity is user 2, then the user and The department relationship feature coefficient of the sender can be A; if the sender’s identity is user 3, the department relationship characteristic coefficient of the user and the sender can be B; if the sender’s identity is user 4, the department of the user and the sender The relation characteristic coefficient can be C.
  • the weight of the minimum unit organization between the user and the sender can also be processed (for example, add a Number or minus one number, etc.), and use the processed data as the characteristic coefficient of the department relationship between the user and the sender.
  • the processed data should meet the above-mentioned setting rules for department relationship characteristic coefficients.
  • Case 7 The user is one of multiple recipients, and the department relationship information of the user includes the smallest unit organization between the user and other users.
  • the department relationship characteristic coefficients include the department relationship characteristic coefficients between the user and the sender, and the department relationship characteristic coefficients between the user and recipients other than the user.
  • the department relationship characteristic coefficients are obtained, including: obtaining the minimum unit organization of the user and the sender from the minimum unit organization of the user and other users, and the minimum unit organization of the user and the recipient other than the user, according to The minimum unit organization between the user and the sender, and the department relationship feature coefficient between the user and the sender is obtained, and the department relationship feature coefficient between the user and the recipient other than the user is obtained according to the minimum unit organization between the user and the recipient other than the user .
  • the specific description of the characteristic coefficient of the department relationship between the user and the recipient other than the user can be referred to in the above case 5.
  • Unit organization get the introduction of the characteristic coefficient of the department relationship between the user and the sender, so I won’t repeat it.
  • Case 8 The user is one of multiple recipients, and the department relationship information of the user includes the weight of the smallest unit organization between the user and other users.
  • the department relationship characteristic coefficients include the department relationship characteristic coefficients between the user and the sender, and the department relationship characteristic coefficients between the user and recipients other than the user.
  • the department relationship feature coefficients are obtained, including: obtaining the weight of the user and the sender’s smallest unit organization from the weight of the user’s and other users’ smallest unit organization, as well as the user’s smallest unit organization weight with the recipient other than the user
  • the weight of the unit organization is based on the weight of the minimum unit organization between the user and the sender, and the department relationship characteristic coefficient between the user and the sender is obtained.
  • the user and the user are obtained The characteristic coefficient of the department relationship of the recipient outside.
  • the specific description of the characteristic coefficients of the department relationship between the user and the recipient other than the user can refer to the above case 6.
  • the weight of the smallest unit organization is introduced by the characteristic coefficients of the department relationship between the user and the sender, and will not be repeated.
  • the characteristic coefficients of the synergy relationship can be obtained in the following four situations:
  • Case 9 The user is the only recipient, and the collaboration relationship information includes indication information that indicates whether the user and other users work together.
  • the characteristic coefficient of the collaboration relationship includes the characteristic coefficient of the collaboration relationship between the user and the sender.
  • Obtaining the characteristic coefficients of the cooperative relationship according to the cooperative relationship information including: obtaining the instruction information indicating whether the user and the sender have cooperative work from the instruction information indicating whether the user and other users have cooperative work, and according to the instruction whether the user and the sender have cooperative work To obtain the characteristic coefficient of the collaborative relationship between the user and the sender.
  • a possible implementation method is to obtain the characteristic coefficients of the cooperative relationship between the user and the sender according to the instruction information indicating whether the user and the sender work together, including: according to the instruction information indicating whether the user and the sender work together.
  • the setting rule of the relationship characteristic coefficient determines the characteristic coefficient of the cooperative relationship between the user and the sender.
  • the coordination relationship feature coefficient setting rule can be as follows: The collaborative relationship feature coefficient when the user has a collaborative relationship with other users is greater than the collaborative relationship feature coefficient when the user has no collaborative relationship with other users. If the collaborative relationship feature coefficients include: the collaborative relationship feature coefficient when the user has no collaborative relationship with other users, the collaborative relationship feature coefficient when the user has a collaborative relationship with other users on important items, and the user has a collaborative relationship with other users on non-important items.
  • the collaborative relationship feature coefficient in the collaborative relationship, and the setting rules for the collaborative relationship feature coefficient can be as follows: the collaborative relationship feature coefficient when the user has a collaborative relationship with other users on important items is larger than the user and other users on non-important items.
  • the characteristic coefficient of the collaborative relationship in the case of a collaborative relationship the characteristic coefficient of the collaborative relationship when the user has a collaborative relationship with other users on non-important items, is greater than the characteristic coefficient of the collaborative relationship when the user has no collaborative relationship with other users.
  • the collaboration relationship information of user 1 includes ⁇ user 1, user 2, 0 ⁇ ; ⁇ user 1, user 3, 1 ⁇ ; ⁇ user 1, user 4, 1 ⁇ , 0 indicates that the user has no collaboration with other users Work, 1 indicates that the user is working in collaboration with other users.
  • the collaborative relationship feature coefficient when the user has a collaborative relationship with other users is 4, and the collaborative relationship feature coefficient when the user has no collaborative relationship with other users is 0.01, for example, if the sender's identity If the user is user 2, the characteristic coefficient of the collaborative relationship between the user and the sender can be 0.01; if the identity of the sender is user 3, the characteristic coefficient of the collaborative relationship between the user and the sender can be 4; if the identity of the sender is user 4, Then the characteristic coefficient of the cooperative relationship between the user and the sender can be 4.
  • Case 10 The user is the only recipient, and the collaboration relationship information includes the weight of the collaboration relationship between the user and other users.
  • the characteristic coefficient of the collaboration relationship includes the characteristic coefficient of the collaboration relationship between the user and the sender.
  • Obtaining the characteristic coefficient of the collaboration relationship according to the collaboration relationship information including: obtaining the weight of the collaboration relationship indicating the user and the sender from the weight of the collaboration relationship between the user and other users, and obtaining the collaboration between the user and the sender according to the weight of the collaboration relationship between the user and the sender Relationship characteristic coefficient.
  • a possible implementation manner is to obtain the characteristic coefficient of the collaborative relationship between the user and the sender according to the weight of the collaborative relationship between the user and the sender, including: taking the weight of the collaborative relationship between the user and the sender as the characteristic coefficient of the collaborative relationship between the user and the sender.
  • the collaborative relationship information of user 1 may include ⁇ user 1, user 2, P ⁇ ; ⁇ user 1, user 3, M ⁇ , where P is the weight of the collaborative relationship between user 1 and user 2, and M is user 1. Take the weight of the collaborative relationship with user 3 as an example. If the sender’s identity is user 2, the characteristic coefficient of the collaborative relationship between the user and the sender can be P; if the sender’s identity is user 3, the collaborative relationship between the user and the sender The characteristic coefficient can be M.
  • the weight of the collaborative relationship between the user and the sender can also be processed (for example, adding a number or subtracting a number). Etc.), and use the processed data as the characteristic coefficient of the collaborative relationship between the user and the sender.
  • the processed data should satisfy the above-mentioned setting rules for the characteristic coefficients of the collaborative relationship.
  • Case 11 The user is one of multiple recipients, and the collaboration relationship information includes indication information indicating whether the user works in collaboration with other users.
  • the collaborative relationship feature coefficients include the collaborative relationship feature coefficients between the user and the sender, and the collaborative relationship feature coefficients between the user and recipients other than the user.
  • Obtaining the characteristic coefficients of the collaborative relationship according to the user's collaborative relationship information including: obtaining the instruction information indicating whether the user and the sender have collaborative work from the instruction information indicating whether the user and other users have collaborative work, and instructing the user to interact with other users According to the instruction information indicating whether the user and the sender have cooperative work, the characteristic coefficient of the cooperative relationship between the user and the sender is obtained, and according to whether the user and the recipient other than the user have cooperation
  • the work instruction information obtains the characteristic coefficient of the collaborative relationship between the user and the recipient other than the user.
  • the specific description of the characteristic coefficient of the cooperative relationship between the user and the sender can be referred to the introduction in the above case 9, and will not be repeated.
  • the indication information indicating whether the user and recipients other than the user have cooperative work the specific description of the characteristic coefficients of the cooperative relationship between the user and the recipients other than the user can refer to the above case 9.
  • the information indicating whether there is cooperative work with the sender, and the introduction of the characteristic coefficients of the cooperative relationship between the user and the sender, will not be repeated.
  • Case 12 The user is one of multiple recipients, and the collaboration relationship information includes the weight of the collaboration relationship between the user and other users.
  • the collaboration relationship feature coefficients include the collaboration relationship feature coefficients between the user and the sender, and the collaboration relationship feature coefficients between the user and recipients other than the user.
  • the collaborative relationship feature coefficients are obtained, including: obtaining the weight of the collaborative relationship between the user and the sender from the weight of the collaborative relationship between the user and other users, and the weight of the collaborative relationship between the user and recipients other than the user, according to The weight of the collaborative relationship between the user and the sender is obtained, and the characteristic coefficient of the collaborative relationship between the user and the sender is obtained.
  • the characteristic coefficient of the collaborative relationship between the user and the recipient other than the user is obtained .
  • the specific description of the characteristic coefficient of the coordination relationship between the user and the sender can be referred to the introduction in the above case 10, and will not be repeated.
  • the weight of the collaborative relationship between the user and the recipient other than the user the specific description of the characteristic coefficient of the collaborative relationship between the user and the recipient other than the user can be referred to in the above case 10.
  • the relationship weight is an introduction to the characteristic coefficients of the collaborative relationship between the user and the sender, and will not be repeated.
  • the characteristic coefficient of the communication relationship is obtained according to the user's communication relationship information and can have the following 4 situations:
  • Case 13 The user is the only recipient, and the user's communication relationship information is used to indicate the frequency of communication between the user and other users.
  • the communication relationship characteristic coefficient includes the communication relationship characteristic coefficient between the user and the sender.
  • the communication relationship characteristic coefficient is obtained based on the user's communication relationship information, including: obtaining the communication frequency between the user and the sender from the user's communication relationship information, and obtaining the communication relationship characteristic coefficient between the user and the sender according to the communication frequency between the user and the sender .
  • a possible implementation method is to obtain the characteristic coefficient of the communication relationship between the user and the sender according to the communication frequency between the user and the sender, including: determining the user and sending according to the communication frequency between the user and the sender and the setting rules of the communication relationship characteristic coefficient Characteristic coefficient of human communication relationship.
  • the communication relationship feature coefficient setting rule may be: the communication relationship feature coefficient is positively related to the communication frequency between the user and the sender, that is, the greater the communication frequency between the user and the sender, the greater the communication relationship feature coefficient.
  • the communication frequency between the user and the sender may be used as the characteristic coefficient of the communication relationship.
  • the user communication relationship information of user 1 includes ⁇ user 1, user 2, 50 ⁇ ; ⁇ user 1, user 3, 30 ⁇ ; ⁇ user 1, user 4, 10 ⁇ , 50 represents the frequency of communication between user 1 and user 2.
  • Is 50 times a day 30 means that the frequency of communication between user 1 and user 3 is 30 times a day
  • 10 means that the frequency of communication between user 1 and user 4 is 10 times a day
  • the sender’s identity is user 2
  • the characteristic coefficient of the communication relationship may be determined according to the correspondence.
  • the correspondence between the communication frequency between the user and the sender and the characteristic coefficient of the communication relationship is shown in Table 5.
  • the communication frequency between the user and the sender is 0-10 times a day
  • the communication relationship characteristic coefficient is 1
  • the communication frequency between the user and the sender is 11-30 times a day
  • the communication relationship characteristic coefficient is 2. If the communication frequency of the sender is 31-50 times a day, the communication relationship characteristic coefficient is 3, and the communication frequency between the user and the sender is more than 50 times a day, and the communication relationship characteristic coefficient is 4.
  • the user communication relationship information of user 1 includes ⁇ user 1, user 2, 50 ⁇ ; ⁇ user 1, user 3, 30 ⁇ ; ⁇ user 1, user 4, 10 ⁇ , 50 represents the frequency of communication between user 1 and user 2. Is 50 times a day, 30 means that the frequency of communication between user 1 and user 3 is 30 times a day, and 10 means that the frequency of communication between user 1 and user 4 is 10 times a day, for example, if the sender’s identity is user 2, then the user and send The characteristic coefficient of a person’s communication relationship can be 3; if the sender’s identity is user 3, the characteristic coefficient of the communication relationship between the user and the sender can be 30; if the sender’s identity is user 2, the communication relationship between the user and the sender The characteristic coefficient can be 1.
  • Table 5 is only an example of the correspondence between the communication frequency between the user and the sender and the characteristic coefficient of the communication relationship.
  • the correspondence between the communication frequency between the user and the sender and the characteristic coefficient of the communication relationship can also be Other forms, and the correspondence between the frequency of communication between the user and the sender and the characteristic coefficients of the communication relationship can be a certain row, some rows, all of Table 5, or more correspondences than those shown in Table 5 , This application is not specifically limited.
  • the frequency of communication between the user and the sender can also be processed (for example, add one number or subtract one number, etc.), and process The latter data is used as the characteristic coefficient of the communication relationship between the user and the sender.
  • the processed data should satisfy the above-mentioned setting rules for the characteristic coefficients of the communication relationship.
  • Case 14 The user is the only recipient, and the user's communication relationship information includes the weight of communication frequency.
  • the communication relationship characteristic coefficient includes the communication relationship characteristic coefficient between the user and the sender.
  • the communication relationship characteristic coefficient is obtained based on the user's communication relationship information, including: obtaining the weight of the communication frequency between the user and the sender from the user's communication relationship information, and obtaining the user and the sender's weight based on the weight of the communication frequency between the user and the sender Characteristic coefficient of communication relationship.
  • a possible implementation method is to obtain the characteristic coefficient of the communication relationship between the user and the sender according to the weight of the communication frequency between the user and the sender, including: using the weight of the communication relationship between the user and the sender as the characteristic coefficient of the communication relationship between the user and the sender .
  • the communication relationship information of user 1 may include ⁇ user 1, user 2, R ⁇ ; ⁇ user 1, user 3, S ⁇ , where R is the weight of the communication relationship between user 1 and user 2, and S is user 1. Take the weight of the communication relationship with user 3 as an example. If the sender’s identity is user 2, the characteristic coefficient of the communication relationship between the user and the sender can be R; if the sender’s identity is user 3, the communication relationship between the user and the sender The characteristic coefficient can be S.
  • the weight of the communication relationship between the user and the sender can also be processed (for example, adding a number or subtracting a number). Etc.), and use the processed data as the characteristic coefficient of the communication relationship between the user and the sender.
  • the processed data should satisfy the above-mentioned setting rules for the characteristic coefficients of the communication relationship.
  • Case 15 The user is one of multiple recipients, and the user's communication relationship information is used to indicate the frequency of communication between the user and other users.
  • the communication relationship characteristic coefficient includes the communication relationship characteristic coefficient between the user and the sender, and the communication relationship characteristic coefficient between the user and the recipient other than the user.
  • the communication relationship characteristic coefficient is obtained based on the user's communication relationship information, including: obtaining the communication frequency between the user and the sender from the user's communication relationship information, and the communication frequency between the user and the receiver other than the user, according to the user and the sender According to the communication frequency between the user and the sender, the characteristic coefficient of the communication relationship between the user and the sender is obtained, and the characteristic coefficient of the communication relationship between the user and the recipient other than the user is obtained according to the communication frequency between the user and the recipient other than the user.
  • the specific description of the characteristic coefficients of the communication relationship between the user and the sender can be referred to the introduction in the above case 13, and will not be repeated.
  • the specific description of the characteristic coefficients of the communication relationship between the user and the recipient other than the user can refer to the case 13 above. , Get the introduction of the characteristic coefficient of the communication relationship between the user and the sender, so I won’t repeat it.
  • Case 16 The user is one of multiple recipients, and the user's communication relationship information includes the weight of the communication frequency.
  • the communication relationship characteristic coefficient includes the communication relationship characteristic coefficient between the user and the sender, and the communication relationship characteristic coefficient between the user and the recipient other than the user.
  • the communication relationship characteristic coefficient is obtained based on the user's communication relationship information, including: the weight of the communication frequency between the user and the sender obtained from the user's communication relationship information, and the weight of the communication frequency between the user and the recipient other than the user, according to The weight of the communication frequency between the user and the sender is obtained, and the characteristic coefficient of the communication relationship between the user and the sender is obtained. According to the weight of the communication frequency between the user and the recipient other than the user, the communication relationship between the user and the recipient other than the user is obtained Characteristic coefficient.
  • the specific description of the characteristic coefficient of the communication relationship between the user and the sender can be referred to the introduction in the above case 14, and will not be repeated.
  • the weight of the communication frequency between the user and the recipients other than the user according to the weight of the communication frequency between the user and the recipients other than the user, the specific description of the characteristic coefficients of the communication relationship between the user and the recipients other than the user can be referred to in the above case 14.
  • the weight of the communication frequency is introduced by the characteristic coefficient of the communication relationship between the user and the sender, and will not be repeated.
  • the relationship feature coefficient can be obtained based on the recipient's identification and/or the sender's identification, and the user's relationship information, but can also be based on the recipient's identification and/or the sender's identification, the user's relationship information, and the second user behavior feedback data get.
  • the second user behavior feedback data can be used to indicate the importance of the text that is the same as the sender of the text in history, and/or the importance of the text that is the same as the recipient of the text in history.
  • the influence coefficient of the second user behavior feedback data can be obtained according to the second user behavior feedback data.
  • the influence coefficient of the second user behavior feedback data can be used Correcting one or more coefficients in the relationship feature coefficients makes the corrected relationship feature coefficients more accurate.
  • the influence coefficient of the second user behavior feedback data is obtained according to the second user behavior feedback data, and one or more coefficients of the relationship feature coefficients obtained by the method described in the above case 1 to case 16 and the second user behavior feedback The influence coefficient of the data is added and subtracted to obtain a more accurate relationship characteristic coefficient.
  • the influence coefficient of the second user behavior feedback data can be used to indicate the importance of the text that is the same as the sender of the text in history, and/or the importance of the text that is the same as the recipient of the text in history.
  • the influence coefficient of the second user behavior feedback data is positively related to the importance of the text, that is, under the condition that other feature coefficients remain unchanged, the greater the influence coefficient of the second user behavior feedback data, the higher the importance of the text .
  • obtaining the influence coefficient of the second user behavior feedback data according to the second user behavior feedback data includes: acquiring a second correspondence, where the second correspondence is the influence coefficient of the second user behavior feedback data and the second user behavior feedback The association relationship between the data determines the influence coefficient of the second user behavior feedback data according to the second user behavior feedback data and the second corresponding relationship.
  • the second correspondence can be preset and stored in the memory 203 or the server in FIG. 2.
  • acquiring the second correspondence relationship includes: acquiring the second correspondence relationship from the memory 203 or the server in FIG. 2.
  • the influence coefficient of the second user behavior feedback data is b.
  • the same text as the sender of the text is non-important text
  • the influence coefficient of the second user behavior feedback data is -b
  • the second user behavior feedback data indicates that the text historically the same as the recipient of the text is important text
  • the influence coefficient of the second user behavior feedback data is c. If the second user behavior feedback data indicates that the text that is the same as the recipient of the text in history is not important, the influence coefficient of the second user behavior feedback data is ⁇ c.
  • the corresponding relationship between the influence coefficient of the second user behavior feedback data and the second user behavior feedback data is shown in Table 6, the upper-lower relationship characteristic coefficient in the relationship characteristic coefficient and the influence coefficient of the second user behavior feedback data Take the addition and subtraction operation as an example.
  • the influence coefficient of the second user behavior feedback data is b, taking the second user behavior feedback into consideration
  • the characteristic coefficient of the subordinate relationship between the new user and the sender can be the characteristic coefficient of the subordinate relationship between the user and the sender + b; if the second user behavior feedback data indicates that it is historically the same as the recipient of the text If the text of is non-important text, the influence coefficient of the second user’s behavior feedback data is -c. After considering the influence coefficient of the second user’s behavior feedback data, the characteristic coefficient of the subordinate relationship between the new user and the recipient can be the user and Recipient's subordinate relationship characteristic coefficient -c.
  • Step 403 Determine the importance of the text according to the relationship feature coefficient.
  • a possible implementation manner is to determine the importance of the text according to the relationship feature coefficient, including: using the relationship feature coefficient as the input data of the machine learning method, and determining the importance of the text through the machine learning method. Specifically, reference may be made to the description of determining the importance of the text through the machine learning method by using the content feature coefficients as the input data of the machine learning method in step 303, which will not be repeated here.
  • the relation feature coefficient when there are multiple texts with the same importance (for example, when there are multiple first-level important texts), after determining the importance of the text according to the relation feature coefficient, it also includes: multiplying the coefficient in the relation feature coefficient Obtain the importance coefficient of the text; sort the text received or sent by the user according to the importance coefficient of the text.
  • textImportantCoff is the importance coefficient of the text
  • upDownOranCoff is the upper-lower relationship characteristic coefficient
  • horizontalCoff is the department relationship characteristic coefficient
  • joinCoff is the synergy relationship characteristic coefficient
  • communicateCoff is the communication relationship characteristic coefficient.
  • the user is one of multiple recipients
  • the upper-lower relationship characteristic coefficient between the user and the sender is upDownOranCoff 1
  • the upper-lower relationship characteristic coefficient between the user and the recipient is upDownOranCoff 2
  • the department relationship between the user and the sender is horizontalCoff 1
  • the department relationship characteristic coefficient between the user and the recipient is horizontalCoff 2
  • the user and sender’s collaborative relationship characteristic coefficient is joinCoff 1
  • the user and the recipient’s collaborative relationship characteristic coefficient is joinCoff 2
  • the characteristic coefficient of the communication relationship is communicateCoff 1
  • the characteristic coefficient of the communication relationship between the user and the recipient is communicateCoff 2.
  • the importance coefficient of the text textImportantCoff upDownOranCoff 1 *horizontalCoff 1 *joinCoff 1 *communicateCoff 1 *upDownOranCoff 2 *horizontalCoff 2 *joinCoff 2 *communicateCoff 2 .
  • Another possible implementation manner is to determine the importance of the text according to the relationship feature coefficients, including: multiplying the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; and determining the importance of the text according to the importance coefficient of the text.
  • determining the importance of the text according to the importance coefficient of the text includes: if the importance coefficient of the text is greater than or equal to the second threshold, determining that the text is an important text; or, according to the importance coefficient of the text and the Correspondence of importance to determine the importance of the text. Specifically, reference may be made to the description of determining the importance of the text according to the importance coefficient of the text in step 303, which is not repeated here.
  • the second threshold and the first threshold may be the same or different.
  • the method further includes: sorting the text received or sent by the user according to the importance coefficient of the text.
  • the text is classified and displayed according to the importance of the text.
  • the method of transmitting the importance of the text in the old device to the new device can refer to the corresponding description in step 303, which will not be repeated here.
  • the recipient’s identity and/or the sender’s identity, and the user’s relationship information can be obtained, and the relationship feature coefficients can be obtained according to the recipient’s identity and/or the sender’s identity and the user’s relationship information.
  • the relationship feature coefficients can be obtained according to the recipient’s identity and/or the sender’s identity and the user’s relationship information.
  • determine the importance of the text according to the relationship feature coefficient so that the importance of the text can be determined according to the relationship between the user and the recipient and/or the sender.
  • the device that recognizes the text can obtain the text, the user's portrait information, and the user's relationship information, and obtain the content feature coefficients according to the text and the user's portrait information, and according to the recipient's identity and/or the sender's identity, and the user
  • the relationship information of the relationship information obtains the relationship feature coefficient, and the importance of the text is determined according to the content feature coefficient and the relationship feature coefficient, so that the importance of the text can be determined according to the content of the text and the relationship between the user and the recipient and/or the sender.
  • the method for recognizing text may include steps 501-504.
  • Step 501 Acquire text, user portrait information, and user relationship information.
  • the text may include the content of the text, the identifier of the recipient and the identifier of the sender.
  • step 501 For the specific introduction of step 501, reference may be made to the corresponding descriptions in step 301 and step 401, which will not be repeated here.
  • Step 502 Obtain content feature coefficients according to the text and the user's portrait information.
  • step 502 For the specific introduction of step 502, reference may be made to the corresponding description in step 302, which is not repeated here.
  • Step 503 Obtain relationship feature coefficients according to the identifier of the recipient and/or the identifier of the sender, and the relationship information of the user.
  • step 503 For the specific introduction of step 503, reference may be made to the corresponding description in step 402, which is not repeated here.
  • step 502 may be executed first and then step 503 may be executed, or step 503 may be executed first and then step 502 may be executed.
  • Step 504 Determine the importance of the text according to the content feature coefficient and the relationship feature coefficient.
  • One possible implementation method is to determine the importance of the text according to the content feature coefficients and the relationship feature coefficients, including: using the content feature coefficients and the relationship feature coefficients as input data of the machine learning method, and determining the importance of the text through the machine learning method. Specifically, reference may be made to the description of determining the importance of the text through the machine learning method by using the content feature coefficients as the input data of the machine learning method in step 303, which will not be repeated here.
  • the text received or sent by the user is sorted according to the importance coefficient of the text.
  • Another possible implementation method is to determine the importance of the text according to the content feature coefficients and the relation feature coefficients, including: adding all or part of the N coefficients in the content feature coefficients, and multiplying them with the coefficients in the relation feature coefficients Obtain the importance coefficient of the text; determine the importance of the text according to the importance coefficient of the text.
  • determining the importance of the text according to the importance coefficient of the text includes: if the importance coefficient of the text is greater than or equal to the third threshold, determining that the text is an important text; or, according to the importance coefficient of the text and the The corresponding relationship of importance determines the importance of the text. Specifically, reference may be made to the description of determining the importance of the text according to the importance coefficient of the text in step 303, which is not repeated here.
  • the third threshold may be the same as or different from the first threshold and the second threshold.
  • the method further includes: sorting the text received or sent by the user according to the importance coefficient of the text.
  • the text is classified and displayed according to the importance of the text.
  • the method of transmitting the importance of the text in the old device to the new device can refer to the corresponding description in step 303, which will not be repeated here.
  • the text, the user’s portrait information, the recipient’s identity and/or the sender’s identity, and the user’s relationship information can be obtained, and the content feature coefficients can be obtained according to the text and the user’s portrait information.
  • the identification of the text and/or the identification of the sender, as well as the user’s relationship information to obtain the relationship feature coefficient, and determine the importance of the text according to the content feature coefficient and the relationship feature coefficient, so that the text content and the user and the recipient and/or sender The relationship between people determines the importance of the text.
  • the above embodiment 1 to embodiment 3 introduce the method of determining the importance of the text according to the content feature coefficient and/or the relationship feature coefficient.
  • the message influence range feature coefficient, and/or the user and the recipient The relationship coefficient between /sender determines the importance of the text.
  • the message influence range characteristic coefficient can be used to indicate the number of recipients
  • the relationship coefficient between the user and the recipient/sender can be used to indicate the closeness of the relationship between the user and the recipient and/or the sender.
  • the following is an example of determining the importance of the text based on the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender.
  • the device that recognizes the text can obtain the text, the user's portrait information, and the user's relationship information, and obtain the content feature coefficients according to the text and the user's portrait information, and according to the recipient's identity and/or the sender's identity, and the user Get the relationship feature coefficient from the relationship information, get the message influence range feature coefficient according to the recipient’s identification, get the relationship coefficient between the user and the receiver/sender according to the user's relationship information, and get the content feature coefficient, relationship feature coefficient, and message influence range
  • the feature coefficient, and the coefficient of the relationship between the user and the recipient/sender determine the importance of the text.
  • the method for recognizing text may include step 601-step 606.
  • Step 601 Obtain the text, the user's portrait information, and the user's relationship information.
  • the text can include the content of the text, the identifier of the recipient, and the identifier of the sender.
  • step 601 For the specific introduction of step 601, reference may be made to the corresponding descriptions in step 301 and step 401, which will not be repeated here.
  • Step 602 Obtain content feature coefficients according to the text and the user's portrait information.
  • step 602 For the specific introduction of step 602, reference may be made to the corresponding description in step 302, which will not be repeated here.
  • Step 603 Obtain relationship feature coefficients according to the identifier of the recipient and/or the identifier of the sender, and the relationship information of the user.
  • step 603 For the specific introduction of step 603, reference may be made to the corresponding description in step 402, which is not repeated here.
  • Step 604 Obtain the characteristic coefficient of the message influence range according to the identifier of the recipient.
  • obtaining the characteristic coefficient of the message influence range according to the identifier of the recipient includes: determining the number of recipients according to the identifier of the recipient, and obtaining the characteristic coefficient of the message influence range according to the number of recipients. For example, if the number of recipients is 2, the message influence range feature coefficient is 2. For example, if the number of recipients is less than 10, the message influence range feature coefficient is 0.3, and if the number of recipients is greater than or equal to 10 and less than 30, the message influence range characteristic coefficient is 0.8, and if the number of recipients is greater than or equal to 30, the message influence range characteristic coefficient is 1.
  • the message influence range feature coefficient is non-positively related to the importance of the text, that is, when other characteristic coefficients remain unchanged, the larger the message influence range feature coefficient, the lower the importance of the text.
  • Step 605 Obtain the relationship coefficient between the user and the receiver/sender according to the relationship information of the user.
  • the relationship coefficient between the user and the recipient/sender can be: the sum of the relationship feature coefficients of the recipients who have an organizational relationship and/or communication relationship with the user, and the sum of the relationship feature coefficients of all recipients Ratio; or, the coefficient of the relationship between the user and the recipient/sender can be: the sum of the weights of the relationship between the recipient and the user who has an organizational and/or communication relationship with the user, and the sum of the weights of the relationship between all recipients and the user ratio.
  • the relationship weight may include the weight of the subordinate relationship, the weight of the smallest unit organization, the weight of the collaboration relationship, and the weight of the communication frequency.
  • relationship feature coefficient of user 2 is 20
  • relationship feature coefficient of user 3 is 33.
  • User 4 has no organizational relationship or communication relationship with user 1.
  • the relationship weight of user 2 is 25
  • the relationship coefficient between the user and the recipient/sender can be: the sum of the relationship characteristic coefficients of the recipient and the sender who have an organizational relationship and/or communication relationship with the user, and the sum of the relationship characteristics of all recipients
  • the ratio of the sum of the characteristic coefficients of the relationship between the person and the sender; or, the relationship coefficient between the user and the recipient/sender can be: the recipient who has an organizational relationship and/or communication relationship with the user and the weight of the relationship between the sender and the user Sum, the ratio to the sum of the weights of all recipients and the relationship between senders and users.
  • user 2 sends text to user 1, user 3, and user 4.
  • User 2 has an organizational relationship with user 1, and the relationship weight of user 2 is 27, and user 3 and user 4 have no organizational relationship with user 1, and user 3
  • the relationship coefficient between the user and the recipient/sender is positively related to the importance of the text, that is, under the condition that other feature coefficients remain unchanged, the greater the relationship coefficient between the user and the recipient/sender, the greater the importance of the text Higher.
  • step 605 may be performed first, then step 604, then step 603, and finally step 602, or step 604 may be performed first. Then step 602 is executed, step 605 is executed, and step 603 is executed finally. It is also possible to execute step 603 first, then execute step 605, then execute step 602, and finally execute step 604.
  • Step 606 Determine the importance of the text according to the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender.
  • a possible implementation method is to determine the importance of the text according to the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender, including: content feature coefficient, relationship feature coefficient, The characteristic coefficients of the influence range of the message and the coefficients of the relationship between the user and the recipient/sender are used as the input data of the machine learning method, and the importance of the text is determined by the machine learning method. Specifically, reference may be made to the description of determining the importance of the text through the machine learning method by using the content feature coefficients as the input data of the machine learning method in step 303, which will not be repeated here.
  • the content feature coefficient, relationship feature coefficient, message influence range feature coefficient, and user and receiver/sender After determining the importance of the text, it also includes: adding all or part of the N coefficients in the content feature coefficients, and multiplying them with the coefficients in the relationship feature coefficients and the relationship coefficients between the user and the recipient/sender. Divide with the feature coefficient of the message influence area to obtain the importance coefficient of the text; sort the text received or sent by the user according to the importance coefficient of the text.
  • the relationship coefficient between the receiveandsendCoff user and the receiver/sender, and textAffectRange is the characteristic coefficient of the message influence range.
  • Another possible implementation method is to determine the importance of the text according to the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the user-recipient/sender relationship coefficient, including: adding N of the content feature coefficients All or part of the coefficients are added together, multiplied with the coefficients in the relationship feature coefficients, the relationship coefficients between the user and the receiver/sender, and divided with the message influence range feature coefficients to get the importance coefficient of the text; according to the importance of the text
  • the sex factor determines the importance of the text.
  • determining the importance of the text according to the importance coefficient of the text includes: if the importance coefficient of the text is greater than or equal to the fourth threshold, determining that the text is an important text; or, according to the importance coefficient of the text and the The corresponding relationship of importance determines the importance of the text. Specifically, reference may be made to the description of determining the importance of the text according to the importance coefficient of the text in step 303, which is not repeated here.
  • the fourth threshold may be the same as or different from the first threshold, the second threshold, and the third threshold.
  • the method further includes: sorting the text received or sent by the user according to the importance coefficient of the text.
  • the text is classified and displayed according to the importance of the text.
  • the method of transmitting the importance of the text in the old device to the new device can refer to the corresponding description in step 303, which will not be repeated here.
  • the text, the user’s portrait information, the recipient’s identity and/or the sender’s identity, and the user’s relationship information can be obtained, and the content feature coefficients can be obtained according to the text and the user’s portrait information.
  • the identity of the sender and/or the identity of the sender, and the relationship information of the user to obtain the relationship characteristic coefficient, the characteristic coefficient of the message influence range according to the identity of the recipient, and the relationship coefficient between the user and the recipient/sender according to the user’s relationship information, and Determine the importance of the text according to the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender.
  • user 1 sends text to user 2 and user 3, and the text recognition device obtains the text, the relationship information of user 2 and the portrait information of user 2, and the content features are obtained according to the text, relationship information of user 2 and portrait information of user 2.
  • Coefficients, relationship feature coefficients, and message influence range feature coefficients are taken as examples to introduce methods for recognizing text.
  • the method for recognizing text includes steps 701-705.
  • Step 701 Obtain the text, the relationship information of the user 2 and the portrait information of the user 2.
  • the text can include the content of the text, the identifier of the recipient, and the identifier of the sender.
  • the relationship information of user 2 includes ⁇ user 1, user 2, superior, small organization 1, 0, 30 ⁇ and ⁇ user 3, user 2, same level, small organization 1, 1, 50 ⁇ .
  • ⁇ user 1, user 2, superior, small organization 1, 0, 30 ⁇ , superior means that user 1 is the superior user of user 2
  • small organization 1 means that the smallest unit organization of user 1 and user 2 is small organization 1
  • 0 means that user 1 and user 2 have no cooperative working relationship
  • 30 means that the communication frequency between user 1 and user 2 is 30 in a month.
  • ⁇ user 3, user 2, same level, small organization 1, 1, 50 ⁇ the same level means that user 3 is a user of the same level of user 2, and small organization 1 means that the smallest unit organization of user 3 and user 2 is small organization 1.
  • 1 indicates that user 3 and user 2 have a cooperative working relationship
  • 50 indicates that the communication frequency between user 3 and user 2 is 50 in a month.
  • the portrait information of user 2 includes ⁇ tag 1 , tag 2 ,..., tag 10 ⁇ and ⁇ tagVal 1 , tagVal 2 ,..., tagVal 10 ⁇ .
  • step 701 For the description of step 701, reference may be made to the introduction in step 301 and step 401, which will not be repeated here.
  • Step 702 Obtain content feature coefficients according to the text and the user's portrait information.
  • step 302 extract the keywords of the text from the text, and obtain the similarity ⁇ similarCoff 1 , similarCoff 2 ,... of the keywords of the text to ⁇ tag 1 , tag 2 ,..., tag 10 ⁇ , SimilarCoff 10 ⁇ .
  • Function(x) exp(x)
  • x tagVal i *similarCoff i
  • i is a positive integer greater than or equal to 1 and less than or equal to 10.
  • Step 703 Obtain relationship feature coefficients according to the identifier of the receiver, the identifier of the sender, and the relationship information of the user.
  • the relationship feature coefficient between user 1 and user 2 can be obtained as ⁇ 6, 5, 0.01, 30 ⁇ , where 6 is the upper and lower relationship feature coefficient of user 1 and user 2, and 5 is user
  • the department relationship feature coefficient between 1 and user 2 0.01 is the collaborative relationship feature coefficient between user 1 and user 2
  • 30 is the communication relationship feature coefficient between user 1 and user 2.
  • the relationship feature coefficient between user 3 and user 2 can be obtained as ⁇ 4, 5, 4, 50 ⁇ , where 4 is the upper and lower relationship feature coefficient of user 3 and user 2, and 5 is user 3 is the department relationship feature coefficient with user 2, 4 is the collaboration relationship feature coefficient between user 3 and user 2, and 50 is the communication relationship feature coefficient between user 3 and user 2.
  • Step 704 Obtain the characteristic coefficient of the message influence range according to the identifier of the recipient.
  • the characteristic coefficient of the message influence range is 2.
  • step 704 the above content feature coefficients, relationship feature coefficients, and message influence range feature coefficients can be combined into a vector [contentCoff 1 , contentCoff 2 ,..., contentCoff 10 , 6, 5 , 0.01, 30, 4, 5 , 4, 50 , 2], a total of 19 dimensions.
  • step 703 may be performed first, then step 702, and finally step 704, or step 704 may be performed first, and then step 702 may be performed. Finally, step 703 is executed, and step 702 may be executed first, then step 704 is executed, and step 703 is executed finally.
  • the embodiment of the present application does not specifically limit it.
  • Step 705 Using content feature coefficients, relationship feature coefficients, and message influence range feature coefficients as input data of the machine learning method, the importance of the text is determined by the machine learning method.
  • step 705 a model of the machine learning method must be trained.
  • the specific process is as follows:
  • Step 1 Obtain a large amount of texts received by user 2 in history. For example, obtain 100000 different texts received by user 2 in history.
  • Step 2 Classify and label the large amount of text according to the importance of the text. For example, important text is marked as 1, and non-important text is marked as 0.
  • Step 3 According to the above steps 701 to 704, obtain the vector (19 dimensions) composed of the content feature coefficients, relationship feature coefficients and message influence range feature coefficients of each text in the large amount of texts (100,000 texts), and The vectors of 100,000 texts are combined in columns to obtain a matrix of 19*100000.
  • Step 4 Use deep learning to train the model.
  • the learning rate is set to 1e-5, and finally the Adam gradient descent optimizer is used to update the parameters, and the number of training is set to 20000 times.
  • the content feature coefficients, relationship feature coefficients, and message influence range feature coefficients of user 2 obtained after step 704 can be combined into a vector [contentCoff 1 , contentCoff 2 ,..., contentCoff 10 , 6, 5 , 0.01, 30, 4, 5, 4, 50, 2] Input the training model to get the importance of the text received by user 2.
  • the text, the user's relationship information, and the user's portrait information can be obtained, and the content feature coefficients can be obtained according to the text and the user's portrait information.
  • the recipient's identity, the sender's identity and the user's relationship information Obtain the relationship feature coefficient, obtain the message influence range characteristic coefficient according to the recipient's identification, and determine the importance of the text according to the content characteristic coefficient, the relationship characteristic coefficient and the message influence range characteristic coefficient, so that the importance of the text can be efficiently recognized.
  • the above-mentioned text recognition apparatus includes hardware structures and/or software modules corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed by hardware or computer software-driven hardware depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
  • the embodiment of the present application may divide the text recognition device into functional modules according to the foregoing method examples.
  • each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or software functional modules. It should be noted that the division of modules in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
  • FIG. 8 shows a schematic structural diagram of a text recognition device 80.
  • the apparatus 80 for recognizing text includes: an acquiring module 801, a processing module 802, and a determining module 803.
  • the obtaining module 801 is used to obtain text and portrait information of a user, where the text is the text received or sent by the user, and the portrait information of the user is used to indicate N keywords in the text related to the user, where N is greater than Or an integer equal to 1.
  • the processing module 802 is configured to obtain a content feature coefficient according to the text and the portrait information of the user, where the content feature coefficient is used to indicate the importance of the content of the text.
  • the determining module 803 is configured to determine the importance of the text according to the content feature coefficient.
  • the portrait information of the user includes N pieces of label information, and the N pieces of label information correspondingly indicate the N keywords; or, the portrait information of the user includes the N pieces of label information, and each of the N pieces of label information The weight of each label information.
  • the processing module 802 is specifically configured to obtain the first user behavior feedback data, where the first user behavior feedback data is used to indicate the importance of a text that is historically similar to the text content; the processing module 802 also It is specifically used to obtain the first correspondence, where the first correspondence is used to indicate the association relationship between the first user behavior feedback data and the influence coefficient of the first user behavior feedback data; the processing module 802 also specifically uses In order to obtain the influence coefficient of the first user behavior feedback data according to the first user behavior feedback data and the first correspondence; the processing module 802 is also specifically configured to obtain the influence coefficient of the first user behavior feedback data according to the text, the portrait information of the user, and the first user behavior The influence coefficient of the feedback data obtains the content characteristic coefficient.
  • the obtaining module 801 is also used to obtain user relationship information, where the user relationship information is used to indicate the hierarchical relationship between the user and other users in the organization; the processing module 802 is also used to obtain information based on the text and the user The relationship information obtained by the relationship feature coefficient, where the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text; the determining module 803 is also used to determine the relationship feature coefficient according to the relationship feature coefficient The importance of text.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration relationship information
  • the user’s subordinate relationship information is used to indicate the user’s subordinate relationship with other users
  • the user’s department relationship information is used to indicate the user’s minimum unit organization with other users, and the minimum unit organization is where the user and other users are located In the same organization, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works in collaboration with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the subordinate relationship characteristic coefficient is based on the The user’s subordinate relationship information is obtained
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the collaboration relationship feature coefficient is obtained based on the collaboration relationship information
  • the communication relationship feature coefficient is based on the user’s communication Relationship information.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient, and the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient;
  • the processing module 802 is specifically used for Acquire second user behavior feedback data, where the second user behavior feedback data is used to indicate the importance of the text that is the same as the sender of the text in history; the processing module 802 is also specifically used to obtain the second correspondence relationship, where: The second corresponding relationship is used to indicate the association relationship between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; the processing module 802 is also specifically configured to indicate the second user behavior feedback data and the second user behavior feedback data. The second corresponding relationship obtains the influence coefficient of the second user behavior feedback data; the processing module 802 is further specifically configured to obtain the relationship characteristic coefficient according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data .
  • the text includes the identifier of the recipient of the text
  • the processing module 802 is further configured to obtain the message influence range characteristic coefficient according to the identifier of the recipient of the text, wherein the message influence range characteristic coefficient is used to indicate the The number of recipients of the text; the determining module 803 is also used to determine the importance of the text according to the characteristic coefficient of the influence range of the message.
  • the processing module 802 is further configured to obtain the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, where the relationship coefficient between the user and the receiver/sender is used to indicate the relationship between the user and the text The degree of closeness of the relationship between the recipient and/or the sender of the text; the determining module 803 is also used to determine the importance of the text according to the relationship coefficient between the user and the recipient/sender.
  • the determining module 803 is specifically configured to use the content feature coefficient as the input data of the machine learning method, and determine the importance of the text through the machine learning method.
  • the determining module 803 is specifically configured to use the content feature coefficient and the relationship feature coefficient as input data of the machine learning method, and determine the importance of the text through the machine learning method.
  • the determining module 803 is specifically configured to use the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient as input data of the machine learning method, and the importance of the text is determined by the machine learning method.
  • the determining module 803 is specifically configured to use the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender as the input data of the machine learning method, and pass the Machine learning methods determine the importance of the text.
  • the content feature coefficients include N coefficients.
  • the device 80 for recognizing text further includes: a sorting module 804; a processing module 802, which is also used to All or part of the N coefficients in the content feature coefficients are added to obtain the importance coefficient of the text; the sorting module 804 is configured to sort the text received or sent by the user according to the importance coefficient of the text.
  • the content feature coefficients include N coefficients.
  • the device 80 for recognizing text further includes: a sorting module 804; a processing module 802, which is also used to The sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients are multiplied to obtain the importance coefficient of the text; the sorting module 804 is configured to perform processing on the text received or sent by the user according to the importance coefficient of the text Sort.
  • the content feature coefficients include N coefficients.
  • the device 80 for recognizing text further includes: a sorting module 804; a processing module 802, which is also used to The sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients are multiplied by the coefficients in the relationship feature coefficients, and divided by the message influence range feature coefficients to obtain the importance coefficient of the text; the sorting module 804 is used for the importance of the text
  • the sexual coefficient ranks the text received or sent by the user.
  • the content feature coefficients include N coefficients.
  • the device 80 for recognizing text further includes: a sorting module 804; a processing module 802, which is also used to The sum of the coefficients in the content feature coefficients is multiplied by the coefficients in the relationship feature coefficients and the relationship between the user and the receiver/sender, and divided by the message influence range feature coefficient to get the importance of the text Coefficient; the sorting module 804 is used to sort the text received or sent by the user according to the importance coefficient of the text.
  • the determining module 803 is specifically configured to add all or part of the N coefficients in the content feature coefficients to obtain the importance coefficient coefficient of the text; the determining module 803 is also specifically configured to perform an addition operation based on the importance of the text The coefficient determines the importance of the text.
  • the determining module 803 is specifically configured to multiply the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; the determining module 803 is also specifically configured to multiply the importance coefficient of the text according to the The importance coefficient of the text determines the importance of the text.
  • the determining module 803 is specifically configured to multiply the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients, and divide the content feature coefficients with the message influence range feature coefficients to obtain the importance coefficient of the text
  • the determination module 803 is also specifically configured to determine the importance of the text according to the importance coefficient of the text.
  • the determining module 803 is specifically configured to multiply the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the relationship coefficients between the user and the recipient/sender, and influence the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text; the determining module 803 is also specifically configured to determine the importance of the text according to the importance coefficient of the text.
  • the sorting module 804 is configured to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the text recognition device 80 further includes: a display module 805; and a display module 805, configured to display the text according to the importance of the text.
  • the text recognition device 80 is presented in the form of dividing various functional modules in an integrated manner.
  • the "module” here can refer to a specific ASIC, circuit, processor and memory that executes one or more software or firmware programs, integrated logic circuit, and/or other devices that can provide the above-mentioned functions.
  • the device 80 for recognizing text may adopt the form shown in FIG. 2.
  • the processor 201 can execute the following process by calling the computer-executed instructions stored in the memory 203: obtain text and user portrait information from the memory 203 or the server, where: The text is the text received or sent by the user, and the portrait information of the user is used to indicate the information related to the importance of the text; the content feature coefficient is obtained according to the text and the portrait information of the user, where the content feature coefficient is To indicate the importance of the content of the text; determine the importance of the text according to the content feature coefficient.
  • the function/implementation process of the acquiring module 801, the processing module 802, the determining module 803, the sorting module 804, and the display module 804 in FIG. 10 can be used to call the computer execution instructions stored in the memory 203 through the processor 201 in FIG. to fulfill.
  • the function/implementation process of the acquiring module 801, the processing module 802, the determining module 803, and the sorting module 804 in FIG. 10 may be implemented by the processor 201 in FIG. 2 calling the computer execution instructions stored in the memory 203, as shown in FIG.
  • the function/implementation process of the display module 805 can be implemented by the output device 205 in FIG. 2.
  • the apparatus 80 for recognizing text provided in this embodiment can perform the above-mentioned method for recognizing text, the technical effects that can be obtained can refer to the above-mentioned method embodiment, which will not be repeated here.
  • FIG. 11 shows a schematic structural diagram of a text recognition device 110.
  • the apparatus 110 for recognizing text includes: an acquiring module 1101, a processing module 1102, and a determining module 1103.
  • the obtaining module 1101 is configured to obtain text and user relationship information, where the text is text received or sent by the user, and the user relationship information is used to indicate the hierarchical relationship between the user and other users in the organization.
  • the processing module 1102 is configured to obtain a relationship feature coefficient according to the text and the relationship information of the user, where the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text.
  • the determining module 1103 is configured to determine the importance of the text according to the relationship feature coefficient.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration relationship information
  • the user’s subordinate relationship information is used to indicate the user’s subordinate relationship with other users
  • the user’s department relationship information is used to indicate the user’s minimum unit organization with other users, and the minimum unit organization is where the user and other users are located In the same organization, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works in collaboration with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the subordinate relationship characteristic coefficient is based on the The user’s subordinate relationship information is obtained
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the collaboration relationship feature coefficient is obtained based on the collaboration relationship information
  • the communication relationship feature coefficient is based on the user’s communication Relationship information.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient, and the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient; the processing module 1102 is specifically used to obtain the first 2.
  • User behavior feedback data where the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history; the processing module 1102 is also specifically used to obtain the second correspondence, where the The second correspondence is used to indicate the correlation between the second user behavior feedback data and the influence coefficient of the second user behavior feedback data; the processing module 1102 is also specifically used to indicate the second user behavior feedback data and the second user behavior feedback data.
  • the processing module 1102 is also specifically configured to obtain the relationship characteristic coefficient according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the determining module 1103 is specifically configured to use the relationship feature coefficient as the input data of the machine learning method, and determine the importance of the text through the machine learning method.
  • the apparatus 110 for recognizing texts further includes: a sorting module 1104; a processing module 1102, which is also used to multiply the coefficients in the relationship feature coefficients The importance coefficient of the text is obtained by calculation; the sorting module 1104 is used to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the determining module 1103 is specifically configured to multiply the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; the determining module 1103 is also specifically configured to determine the importance of the text according to the importance coefficient of the text. Sex.
  • the sorting module 1104 is further configured to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the apparatus 110 for recognizing text further includes: a display module 1105; a display module 1105, configured to display the text according to the importance of the text.
  • the device 110 for recognizing text is presented in the form of dividing various functional modules in an integrated manner.
  • the "module” here can refer to a specific ASIC, circuit, processor and memory that executes one or more software or firmware programs, integrated logic circuit, and/or other devices that can provide the above-mentioned functions.
  • the device 110 for recognizing text may adopt the form shown in FIG. 2.
  • the processor 201 can execute the following process by calling the computer-executable instructions stored in the memory 203: obtaining text and user relationship information from the memory 203 or the server, where: The text is the text received or sent by the user, and the relationship information of the user is used to indicate the hierarchical relationship between the user and other users in the organization; the relationship feature coefficient is obtained according to the text and the relationship information of the user, where the relationship feature coefficient is To indicate the relationship between the user and the recipient of the text and/or the sender of the text; determine the importance of the text according to the relationship feature coefficient.
  • the function/implementation process of the acquiring module 1101, the processing module 1102, the determining module 1103, the sorting module 1104, and the display module 1104 in FIG. 13 can be used to call the computer execution instructions stored in the memory 203 through the processor 201 in FIG. to fulfill.
  • the function/implementation process of the acquiring module 1101, the processing module 1102, the determining module 1103, and the sorting module 1104 in FIG. 13 can be implemented by the processor 201 in FIG. 2 calling the computer execution instructions stored in the memory 203, in FIG.
  • the function/implementation process of the display module 1105 can be implemented by the output device 205 in FIG. 2.
  • the apparatus 110 for recognizing text provided in this embodiment can execute the above-mentioned method for recognizing text, the technical effects that can be obtained can refer to the above-mentioned method embodiment, and will not be repeated here.
  • FIG. 14 shows a schematic structural diagram of a text recognition device 140.
  • the apparatus 140 for recognizing text includes: an acquiring module 1401, a processing module 1402, and a determining module 1403.
  • the obtaining module 1401 is used to obtain text, user portrait information, and user relationship information, where the text is the text received or sent by the user, and the user portrait information is used to indicate N of the user-related text Keyword, N is an integer greater than or equal to 1, and the relationship information of the user is used to indicate the hierarchical relationship between the user and other users in the organization.
  • the processing module 1402 is configured to obtain a content feature coefficient according to the text and the portrait information of the user, where the content feature coefficient is used to indicate the importance of the content of the text.
  • the processing module 1402 is further configured to obtain a relationship feature coefficient according to the text and the relationship information of the user, where the relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text.
  • the determining module 1403 is configured to determine the importance of the text according to the content feature coefficient and the relationship feature coefficient.
  • the portrait information of the user includes N pieces of label information, and the N pieces of label information correspondingly indicate the N keywords; or, the portrait information of the user includes the N pieces of label information, and each of the N pieces of label information The weight of each label information.
  • the processing module 1402 is specifically configured to obtain first user behavior feedback data, where the first user behavior feedback data is used to indicate the importance of a text with a high similarity to the text content in history; the processing module 1402 also It is specifically used to obtain the first correspondence, where the first correspondence is used to indicate the association relationship between the first user behavior feedback data and the influence coefficient of the first user behavior feedback data; the processing module 1402 also specifically uses In order to obtain the influence coefficient of the first user behavior feedback data according to the first user behavior feedback data and the first corresponding relationship; the processing module 1402 is also specifically configured to obtain the influence coefficient of the first user behavior feedback data according to the text, the portrait information of the user, and the first user The influence coefficient of the behavior feedback data obtains the content characteristic coefficient.
  • the user's relationship information includes the user's organizational relationship information and the user's communication relationship information
  • the user's organizational relationship information includes the user's superior-subordinate relationship information, the user's department relationship information, and collaboration relationship information
  • the user’s subordinate relationship information is used to indicate the user’s subordinate relationship with other users
  • the user’s department relationship information is used to indicate the user’s minimum unit organization with other users, and the minimum unit organization is where the user and other users are located In the same organization, the organization with the least number of users
  • the collaboration relationship information is used to indicate whether the user works in collaboration with other users
  • the communication relationship information of the user is used to indicate the frequency of communication between the user and other users.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient
  • the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient
  • the subordinate relationship characteristic coefficient is based on the The user’s subordinate relationship information is obtained
  • the department relationship feature coefficient is obtained based on the user’s department relationship information
  • the collaboration relationship feature coefficient is obtained based on the collaboration relationship information
  • the communication relationship feature coefficient is based on the user’s communication Relationship information.
  • the relationship characteristic coefficient includes an organization relationship characteristic coefficient and a communication relationship characteristic coefficient, and the organization relationship characteristic coefficient includes a subordinate relationship characteristic coefficient, a department relationship characteristic coefficient, and a synergy relationship characteristic coefficient;
  • the processing module 1402 is specifically used to obtain the first 2.
  • User behavior feedback data where the second user behavior feedback data is used to indicate the importance of the same text as the sender of the text in history; the processing module 1402 is also specifically used to obtain the second correspondence relationship, where the first The second correspondence is used to indicate the second user behavior feedback data and the association relationship between the influence coefficients of the second user behavior feedback data; the processing module 1402 is also specifically used to indicate the second user behavior feedback data and the second user behavior feedback data.
  • Two corresponding relationships to obtain the influence coefficient of the second user behavior feedback data; the processing module 1402 is further specifically configured to obtain the relationship characteristic coefficient according to the text, the relationship information of the user, and the influence coefficient of the second user behavior feedback data.
  • the processing module 1402 is further configured to obtain a characteristic coefficient of the influence range of the message according to the text, wherein the characteristic coefficient of the influence range of the message is used to indicate the number of recipients of the text; the determining module 1403 is also used to The characteristic coefficient of the influence range of the message determines the importance of the text.
  • the processing module 1402 is further configured to obtain the relationship coefficient between the user and the receiver/sender according to the relationship information of the user, wherein the relationship coefficient between the user and the receiver/sender is used to indicate the relationship between the user and the The degree of closeness of the relationship between the recipient of the text and/or the sender of the text; the determining module 1403 is also used to determine the importance of the text according to the relationship coefficient between the user and the recipient/sender.
  • the determining module 1403 is specifically configured to use the content feature coefficient and the relationship feature coefficient as input data of the machine learning method, and determine the importance of the text through the machine learning method.
  • the determining module 1403 is specifically configured to use the content feature coefficient, the relationship feature coefficient, and the message influence range feature coefficient as input data of the machine learning method, and determine the importance of the text through the machine learning method.
  • the determining module 1403 is specifically configured to use the content feature coefficient, the relationship feature coefficient, the message influence range feature coefficient, and the relationship coefficient between the user and the recipient/sender as the input data of the machine learning method.
  • Machine learning methods determine the importance of the text.
  • the apparatus 140 for recognizing text further includes: a sorting module 1404; a processing module 1402, which is also used for Multiply the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; the sorting module 1404 is used for the text received or sent by the user according to the importance coefficient of the text put in order.
  • the apparatus 140 for recognizing text further includes: a sorting module 1404; a processing module 1402, which is also used for Multiply the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients, and divide them with the message’s influence range feature coefficients to obtain the importance coefficient of the text; the sorting module 1404 is used to calculate the text according to the The importance factor ranks the text received or sent by the user.
  • the apparatus 140 for recognizing text further includes: a sorting module 1404; a processing module 1402, which is also used for Multiply the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the relationship coefficients between the user and the recipient/sender, and divide it with the message influence range feature coefficient to get the importance of the text Sexual coefficient; a sorting module 1404, used to sort the text received or sent by the user according to the importance coefficient of the text.
  • the determining module 1403 is specifically configured to multiply the sum of the coefficients in the content feature coefficients and the coefficients in the relationship feature coefficients to obtain the importance coefficient of the text; the determining module 1403 is also specifically configured to multiply the importance coefficient of the text according to the The importance coefficient of the text determines the importance of the text.
  • the determining module 1403 is specifically configured to multiply the sum of the coefficients in the content feature coefficients with the coefficients in the relationship feature coefficients, and divide with the feature coefficients of the influence range of the message to obtain the importance coefficient of the text ;
  • the determining module 1403 is also specifically configured to determine the importance of the text according to the importance coefficient of the text.
  • the determining module 1403 is specifically configured to multiply the sum of the coefficients in the content feature coefficients, the coefficients in the relationship feature coefficients, and the relationship coefficients between the user and the recipient/sender, and influence the message
  • the range feature coefficient is divided to obtain the importance coefficient of the text; the determining module 1403 is also specifically configured to determine the importance of the text according to the importance coefficient of the text.
  • the sorting module 1404 is further configured to sort the texts received or sent by the user according to the importance coefficient of the text.
  • the apparatus 140 for recognizing text further includes: a display module 1405; the display module 1405 is configured to display the text according to the importance of the text.
  • the device 140 for recognizing text is presented in the form of dividing various functional modules in an integrated manner.
  • the "module” here can refer to a specific ASIC, circuit, processor and memory that executes one or more software or firmware programs, integrated logic circuit, and/or other devices that can provide the above-mentioned functions.
  • the device 140 for recognizing text may adopt the form shown in FIG. 2.
  • the processor 201 can execute the following process by calling the computer-executable instructions stored in the memory 203: obtaining text, user portrait information, and user information from the memory 203 or the server
  • the relationship information of the user where the text is the text received or sent by the user, the portrait information of the user is used to indicate information related to the importance of the text, and the relationship information of the user is used to indicate the relationship between the user and other users in the organization Hierarchical relationship; obtain the content feature coefficient according to the text and the user's portrait information, where the content feature coefficient is used to indicate the importance of the content of the text; obtain the relationship feature coefficient according to the text and the relationship information of the user, where the The relationship feature coefficient is used to indicate the relationship between the user and the recipient of the text and/or the sender of the text.
  • the importance of the text is determined according to a feature coefficient set, where the feature coefficient set includes the content feature coefficient and the relationship feature coefficient.
  • the function/implementation process of the acquiring module 1401, the processing module 1402, the determining module 1403, the sorting module 1404, and the display module 1405 in FIG. 16 may call the computer execution instructions stored in the memory 203 through the processor 201 in FIG. to fulfill.
  • the function/implementation process of the acquiring module 1401, the processing module 1402, the determining module 1403, and the sorting module 1404 in FIG. 16 can be implemented by the processor 201 in FIG. 2 calling the computer execution instructions stored in the memory 203, in FIG.
  • the function/implementation process of the display module 1405 can be implemented by the output device 205 in FIG. 2.
  • the apparatus 140 for recognizing text provided in this embodiment can perform the above-mentioned method for recognizing text, the technical effects that can be obtained can refer to the above-mentioned method embodiment, and will not be repeated here.
  • the computer may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
  • a software program it may be implemented in the form of a computer program product in whole or in part.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • the computer instructions may be transmitted from a website, computer, server, or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer, or may include one or more data storage devices such as servers and data centers that can be integrated with the medium.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Computational Linguistics (AREA)
  • Business, Economics & Management (AREA)
  • Artificial Intelligence (AREA)
  • Human Resources & Organizations (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mathematical Physics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Strategic Management (AREA)
  • Multimedia (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Document Processing Apparatus (AREA)

Abstract

一种识别文本的方法及装置，涉及信息分类领域，可以获取文本和用户的画像信息（301），根据文本和用户的画像信息得到内容特征系数（302），并根据内容特征系数确定文本的重要性（303）。如此，可以在大量的文本中，根据文本的内容自动筛选出与用户相关的文本，效率较高。

Description

识别文本的方法及装置
申请要求在2019年09月30日提交中国国家知识产权局、申请号为201910944108.9的中国专利申请的优先权，发明名称为“识别文本的方法及装置”的中国专利申请的优先权，在2019年08月02日提交中国国家知识产权局、申请号为201910712935.5的中国专利申请的优先权，发明名称为“识别用户重要短文本的方法和计算机设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及信息分类领域，尤其涉及识别文本的方法及装置。
背景技术
随着组织型结构领域(例如：面向企业(to business，2B)领域)中移动办公的兴起，移动办公软件在人们的工作中的作用越来越重要，人们在工作中可以通过移动办公软件及时沟通，从而提高工作效率。
目前，对于使用移动办公软件的用户而言，每天接收或发送的文本(例如：即时聊天内容、邮件、会议通知等)呈爆炸性增长。但是，用户接收或发送的的文本中，仅有部分文本是与该用户相关的，因此，该用户在接收或发送文本后，需要手动筛选与自身相关的文本。然而，在大量的文本中，手动筛选与自身相关的文本，不仅耗时长而且效率低。
发明内容
本申请实施例提供识别文本的方法及装置，可以识别用户接收或者发送的文本的重要性。
为达到上述目的，本申请实施例采用如下技术方案：
第一方面，本申请实施例提供一种识别文本的方法，该方法包括：获取文本以及用户的画像信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示与该用户相关的文本中的N个关键词，N为大于或等于1的整数；根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示该文本的内容的重要性；根据该内容特征系数确定该文本的重要性。
上述第一方面提供的技术方案，可以获取文本以及用户的画像信息，根据文本以及用户的画像信息得到内容特征系数，并根据该内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的画像信息包括N个标签信息，该N个标签信息对应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。基于上述技术方案，可以根据文本以及N个标签信息得到内容特征系数，并根据内容特征系数确定文本的重要性，或者，可以根据文本、N个标签信息以及N个标签信息中每个标签信息的权重得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该内容特征系数包括N个系数，其中，第n个系数根据该文本和该用户的画像信息中第n个标签信息得到，n为大于0且小于或等于N的正整数。 基于上述技术方案，可以根据文本和用户画像信息中第n个标签信息得到内容特征系数中的第n个系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，根据该文本以及该用户的画像信息得到内容特征系数，包括：获取第一用户行为反馈数据，其中，该第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；获取第一对应关系，其中，该第一对应关系用于指示该第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；根据该第一用户行为反馈数据和该第一对应关系得到该第一用户行为反馈数据的影响系数；根据该文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到该内容特征系数。基于上述技术方案，可以获取第一用户行为反馈数据，根据第一用户行为反馈数据以及第一关系得到第一用户行为反馈数据的影响系数，根据文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该方法还包括：获取用户的关系信息，其中，该用户的关系信息用于指示用户与其他用户的层级关系；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；该确定该文本的重要性，还包括：根据该关系特征系数确定该文本的重要性。基于上述技术方案，可以获取用户的关系信息，根据文本以及用户的关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据内容特征系数、上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该根据该文本以及该用户的关系信息得到关系特征系数，包括：获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该文本包括该文本的接收人的标识，该方法还包括：根据该文本的接收人的标识得到该消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数；该确定该文本的重要性，还包括：根据该消息影响范围特征系数确定该文本的重要性。基于上述技术方案，可以根据该文本得到消息影响范围特征系数，根据内容特征系数、该关系特征系数以及消息影响范围特征系数确定文本的 重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及文本的接收人的个数识别用户接收或发送的文本的重要性。
一种可能的实现方式，该方法还包括：根据该用户的关系信息得到用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度；该确定该文本的重要性，还包括：根据该用户与接收人/发送人的关系系数确定该文本的重要性。基于上述技术方案，可以根据用户的关系信息得到用户与接收人/发送人的关系系数，根据内容特征系数、关系特征系数以及用户与接收人/发送人的关系系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及用户与文本的接收人和/或文本的发送人的关系的紧密程度识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：以该内容特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到内容特征系数后，可以将内容特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：以该内容特征系数和该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到内容特征系数和关系特征系数后，可以将内容特征系数和关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数、该消息影响范围特征系数以及该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到内容特征系数、关系特征系数、消息影响范围特征系数以及用户与接收人/发送人的关系系数后，可以将内容特征系数、关系特征系数、消息影响范围特征系数以及用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该方法还包括：将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的N个系数全部或部分做加法运算得到文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该方法还包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该方法还包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该方法还包括：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数后，可以将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数和关系特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要 性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性系数。基于上述技术方案，在得到该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数后，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该方法还包括：根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该方法还包括：将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第二方面，本申请实施例提供一种识别文本的方法，该方法包括：获取文本以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的关系信息用于指示该用户与其他用户的层级关系；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；根据该关系特征系数确定该文本的重要性。
上述第二方面提供的技术方案，可以获取文本以及用户的关系信息，根据文本以及用户的关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关 系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该根据该文本以及该用户的关系信息得到关系特征系数，包括：获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据该关系特征系数确定该文本的重要性，包括：以该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到关系特征系数后，可以将关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该方法还包括：将该关系特征系数中的系数做乘法运算得到该文本的重要性系数系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将关系特征系数中的系数做乘法运算得到文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该根据该关系特征系数确定该文本的重要性，包括：将该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到关系特征系数后，可以将关系特征系数中的系数做乘法运算得到文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该方法还包括：根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该方法还包括：将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第三方面，本申请实施例提供一种识别文本的方法，该方法包括：获取文本、用户的画像信息以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示与该用户相关的文本中的N个关键词，N为大于或等于1的整数，该用户的关系信息用于指示该用户与其他用户的层级关系；根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示文本的内容的重要性；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；根据该内容特征系数以及该关系特征系数确定该文本的重要性。
上述第三方面提供的技术方案，可以获取文本、用户的画像信息以及用户的关系信息，根据文本以及用户的画像信息得到内容特征系数，根据文本以及用户的关系信息得到关系特征系数，并根据内容特征系数和关系特征系数确定该文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的画像信息包括N个标签信息，该N个标签信息对应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。基于上述技术方案，可以根据文本以及N个标签信息得 到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，或者，可以根据文本、N个标签信息以及N个标签信息中每个标签信息的权重得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该内容特征系数包括N个系数，其中，第n个系数根据该文本和该用户的画像信息中第n个标签信息得到，n为大于0且小于或等于N的正整数。基于上述技术方案，可以根据文本和用户画像信息中第n个标签信息得到内容特征系数中的第n个系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，根据该文本以及该用户的画像信息得到内容特征系数，包括：获取第一用户行为反馈数据，其中，第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；获取第一对应关系，其中，该第一对应关系用于指示该第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；根据该第一用户行为反馈数据和该第一对应关系得到该第一用户行为反馈数据的影响系数；根据该文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到该内容特征系数。基于上述技术方案，可以获取第一用户行为反馈数据，根据第一用户行为反馈数据以及第一关系得到第一用户行为反馈数据的影响系数，根据文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据内容特征系数、上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该根据该文本以及该用户的关系信息得到关系特征系数，包括：获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该方法还包括:根据该文本得到该消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数；该确定该文本的重要性，还包括：根据该消息影响范围特征系数确定该文本的重要性。基于上述技术方案，可以根据该文本得到消息影响范围特征系数，根据内容特征系数、关系特征系数以及消息影响范围特征系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和 发送人的关系以及文本的接收人的个数识别用户接收或发送的文本的重要性。
一种可能的实现方式，该方法还包括：根据该用户的关系信息得到该用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度；该确定该文本的重要性，还包括：根据该用户与接收人/发送人的关系系数确定该文本的重要性。基于上述技术方案，可以根据用户的关系信息得到用户与接收人/发送人的关系系数，根据内容特征系数，关系特征系数以及用户与接收人/发送人的关系系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及用户与文本的接收人和/或文本的发送人的关系的紧密程度识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：以该内容特征系数以及该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到特征系数集合后，可以将内容特征系数以及关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到特征系数集合后，可以将内容特征系数、关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到特征系数集合后，可以将内容特征系数、关系特征系数和该消息影响范围特征系数和该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该方法还包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该方法还包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特 征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该方法还包括：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数和关系特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数后，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该方法还包括：根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该方法还包括：将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第四方面，本申请实施例提供一种识别文本的装置，该装置包括：获取模块、处理模块以及确定模块；该获取模块，用于获取文本以及用户的画像信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示该用户相关的文本中的N个关键词，N为大于或等于1的整数；该处理模块，用于根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示该文本的内容的重要性；该确定模块，用于根据内容特征系数确定该文本的重要性。
上述第四方面提供的技术方案，可以获取文本以及用户的画像信息，根据文本以及用户的画像信息得到内容特征系数，并根据该内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的画像信息包括N个标签信息，该N个标签信息对应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。基于上述技术方案，可以根据文本以及N个标签信息得到内容特征系数，并根据内容特征系数确定文本的重要性，或者，可以根据文本、N个标签信息以及N个标签信息中每个标签信息的权重得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该内容特征系数包括N个系数，其中，第n个系数根据该文本和该用户的画像信息中第n个标签信息得到，n为大于0且小于或等于N的正整数。基于上述技术方案，可以根据文本和用户画像信息中第n个标签信息得到内容特征系数中的第n个系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该处理模块，具体用于获取第一用户行为反馈数据，其中，第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；该处理模块，还具体用于获取第一对应关系，其中，该第一对应关系用于指示该第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；该处理模块，还具体用于根据该第一用户行为反馈数据和该第一对应关系得到该第一用户行为反馈数据的影响系数；该处理模块，还具体用于根据该文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到该内容特征系数。基于上述技术方案，可以获取第一用户行为反馈数据，根据第一用户行为反馈数据以及第一关系得到第一用户行为反馈数据的影响系数，根据文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该获取模块，还用于获取用户的关系信息，其中，该用户的关系信息用于指示用户与其他用户的层级关系；该处理模块，还用于根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系，该确定模块，还用于根据该关系特征系数确定该文 本的重要性。基于上述技术方案，可以获取用户的关系信息，根据文本以及用户的关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通 关系信息得到沟通关系特征系数，并根据内容特征系数、上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该处理模块，具体用于用于获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；该处理模块，还具体用于根据该第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；该处理模块，还具体用于根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该文本包括该文本的接收人的标识，该处理模块，还用于根据该文本的接收人的标识得到该消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数，该确定模块，还用于根据该消息影响范围特征系数确定该文本的重要性。基于上述技术方案，可以根据该文本得到消息影响范围特征系数，根据内容特征系数、该关系特征系数以及消息影响范围特征系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及文本的接收人的个数识别用户接收或发送的文本的重要性。
一种可能的实现方式，该处理模块，还用于根据该用户的关系信息得到用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度，该确定模块，还用于根据该用户与接收人/发送人的关系系数确定该文本的重要性。基于上述技术方案，可以根据用户的关系信息得到用户与接收人/发送人的关系系数，根据内容特征系数、关系特征系数以及用户与接收人/发送人的关系系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及用户与文本的接收人和/或文本的发送人的关系的紧密程度识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该内容特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数后，可以将该内容特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该内容特征系数和该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到内容特征系数和关系特征系数后，可以将内容特征系数和关系特征系数 作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该内容特征系数、该关系特征系数、该消息影响范围特征系数以及该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到内容特征系数、关系特征系数、消息影响范围特征系数以及用户与接收人/发送人的关系系数后，可以将内容特征系数、关系特征系数、消息影响范围特征系数以及用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该装置还包括：排序模块；该处理模块，还用于将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该装置还包括：排序模块；该处理模块，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该装置还包括：排序模块；该处理模块，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该装置还包括：排序模块；该处理模块，还用于将该内容特征系数中的系数之和， 与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该确定模块，具体用于将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数后，可以将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数和该关系特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数后，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该装置还包括：排序模块；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该装置还包括：显示模块；该显示模块，用于将该文本按照 该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第五方面，本申请实施例提供一种识别文本的装置，该装置包括：获取模块、处理模块以及确定模块；该获取模块，用于获取文本以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的关系信息用于指示该用户与其他用户的层级关系；该处理模块，用于根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；该确定模块，用于根据该关系特征系数确定该文本的重要性。
上述第五方面提供的技术方案，可以获取文本以及用户的关系信息，根据文本以及用户的关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系 数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该处理模块，具体用于获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；该处理模块，还具体用于获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据，以及该第二用户行为反馈数据的影响系数之间的关联关系；该处理模块，还具体用于根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；该处理模块，还具体用于根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到关系特征系数后，可以将关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该装置还包括：排序模块；该处理模块，还用于将该关系特征系数中的系数做乘法运算得到该文本的重要性系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将关系特征系数中的系数做乘法运算得到文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该确定模块，具体用于将该关系特征系数中的系数做乘法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到关系特征系数后，可以将关系特征系数中的系数做乘法运算得到文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该装置还包括：排序模块；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于 上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该装置还包括：显示模块；该显示模块，用于将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第六方面，本申请实施例提供一种文本识别的装置，该装置包括：获取模块、处理模块以及确定模块；该获取模块，用于获取文本、用户的画像信息、以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示该用户相关的文本中的N个关键词，N为大于或等于1的整数，该用户的关系信息用于指示该用户与其他用户的层级关系；该处理模块，用于根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示文本的内容的重要性；该处理模块，还用于根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；该确定模块，用于根据该内容特征系数以及该关系特征系数确定该文本的重要性。
上述第六方面提供的技术方案，可以获取文本、用户的画像信息以及用户的关系信息，根据文本以及用户的画像信息得到内容特征系数，根据文本以及用户的关系信息得到关系特征系数，并根据内容特征系数和关系特征系数确定该文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的画像信息包括N个标签信息，该N个标签信息对应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。基于上述技术方案，可以根据文本以及N个标签信息得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，或者，可以根据文本、N个标签信息以及N个标签信息中每个标签信息的权重得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该内容特征系数包括N个系数，其中，第n个系数根据该文本和该用户的画像信息中第n个标签信息得到，n为大于0且小于或等于N的正整数。基于上述技术方案，可以根据文本和用户画像信息中第n个标签信息得到内容特征系数中的第n个系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该处理模块，具体用于获取第一用户行为反馈数据，其中，第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；该处理模块，还具体用于获取第一对应关系，其中，该第一对应关系用于指示该第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；该处理模块，还具体用于根据该第一用户行为反馈数据和该第一对应关系得到该第一用户行为反馈数据的影响系数；该处理模块，还具体用于根据该文本、该用户的画像信息以及该第一用户行 为反馈数据的影响系数得到该内容特征系数。基于上述技术方案，可以获取第一用户行为反馈数据，根据第一用户行为反馈数据以及第一关系得到第一用户行为反馈数据的影响系数，根据文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方 案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据内容特征系数、上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该处理模块，具体用于获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；该处理模块，还具体用于获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；该处理模块，还具体用于根据该第二用户行为反馈数据以及该第二对应关系，该得到第二用户行为反馈数据的影响系数；该处理模块，还具体用于根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该处理模块，还用于根据该文本得到该消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数，该确定模块，还用于根据该消息影响范围特征系数确定该文本的重要性。基于上述技术方案，可以根据该文本得到消息影响范围特征系数，根据内容特征系数、关系特征系数以及消息影响范围特征系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及文本的接收人的个数识别用户接收或发送的文本的重要性。
一种可能的实现方式，该处理模块，还用于根据该用户的关系信息得到该用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度，该确定模块，还用于根据该用户与接收人/发送人的关系系数确定该文本的重要性。基于上述技术方案，可以根据用户的关系信息得到用户与接收人/发送人的关系系数，根据内容特征系数，关系特征系数以及用户与接收人/发送人的关系系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及用户与文本的接收人和/或文本的发送人的关系的紧密程度识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该内容特征系数以及该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数以及该关系特征系数后，可以将该内容特征系数以及该关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该内容特征系数、该关系特征系数 和该消息影响范围特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于以该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数后，可以将该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该装置还包括：排序模块；该处理模块，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该装置还包括：排序模块；该处理模块，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该装置还包括：排序模块；该处理模块，还用于将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该确定模块，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征 系数和该关系特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该确定模块，具体用于将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；该确定模块，还具体用于根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数后，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该装置还包括：排序模块；该排序模块，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该装置还包括：显示模块；该显示模块，用于将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第七方面，本申请提供了一种识别文本的装置，该装置可以包括：至少一个处理器、存储器，该存储器存储有软件程序，该处理器用于调用该存储器中的软件程序执行下述过程：从该存储器或服务器中获取文本以及用户的画像信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示该用户相关的文本中的N个关键词，N为大于或等于1的整数，该服务器与该识别文本的装置连接，该服务器中存储有该文本以及该用户的画像信息；根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示该文本的内容的重要性；根据该内容特征系数确定该文本的重要性。
上述第七方面提供的技术方案，可以获取文本以及用户的画像信息，根据文本以及用户的画像信息得到内容特征系数，并根据该内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的画像信息包括N个标签信息，该N个标签信息对 应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。基于上述技术方案，可以根据文本以及N个标签信息得到内容特征系数，并根据内容特征系数确定文本的重要性，或者，可以根据文本、N个标签信息以及N个标签信息中每个标签信息的权重得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该内容特征系数包括N个系数，其中，第n个系数根据该文本和该用户的画像信息中第n个标签信息得到，n为大于0且小于或等于N的正整数。基于上述技术方案，可以根据文本和用户画像信息中第n个标签信息得到内容特征系数中的第n个系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该根据该文本以及该用户的画像信息得到内容特征系数，包括：从该存储器或该服务器中获取第一用户行为反馈数据，其中，第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；从该存储器或该服务器中获取第一对应关系，其中，该第一对应关系用于指示该第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；根据该第一用户行为反馈数据和该第一对应关系得到该第一用户行为反馈数据的影响系数；根据该文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到该内容特征系数。基于上述技术方案，可以获取第一用户行为反馈数据，根据第一用户行为反馈数据以及第一关系得到第一用户行为反馈数据的影响系数，根据文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该处理器还用于调用该存储器中的软件程序执行下述过程：从该存储器或该服务器中获取用户的关系信息，其中，该用户的关系信息用于指示用户与其他用户的层级关系；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；该确定该文本的重要性，还包括：根据该关系特征系数确定该文本的重要性。基于上述技术方案，可以获取用户的关系信息，根据文本以及用户的关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户 与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据内容特征系数、上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该根据该文本以及该用户的关系信息得到关系特征系数，包括：从该存储器或该服务器中获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据，以及该第二用户行为反馈数据的影响系数之间的关联关系；根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第 二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该文本包括该文本的接收人的标识；该处理器还用于调用该存储器中的软件程序执行下述过程：根据该文本的接收人的标识得到消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数；该确定该文本的重要性，还包括：根据该消息影响范围特征系数确定该文本的重要性。基于上述技术方案，可以根据该文本得到消息影响范围特征系数，根据内容特征系数、该关系特征系数以及消息影响范围特征系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及文本的接收人的个数识别用户接收或发送的文本的重要性。
一种可能的实现方式，该处理器还用于调用该存储器中的软件程序执行下述过程：根据该用户的关系信息得到用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度；该确定该文本的重要性，还包括：根据该用户与接收人/发送人的关系系数确定该文本的重要性。基于上述技术方案，可以根据用户的关系信息得到用户与接收人/发送人的关系系数，根据内容特征系数、关系特征系数以及用户与接收人/发送人的关系系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及用户与文本的接收人和/或文本的发送人的关系的紧密程度识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据该内容特征系数确定该文本的重要性，包括：以该内容特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数后，可以将该内容特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：以该内容特征系数和该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到内容特征系数和关系特征系数后，可以将内容特征系数和关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数、该消息影响范围特征系数以及该用户与接收人/发送人的关 系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到内容特征系数、关系特征系数、消息影响范围特征系数以及用户与接收人/发送人的关系系数后，可以将内容特征系数、关系特征系数、消息影响范围特征系数以及用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该处理器还用于调用该存储器中的软件程序执行下述过程：将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的N个系数全部或部分做加法运算得到文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该处理器还用于调用该存储器中的软件程序执行下述过程：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该处理器还用于调用该存储器中的软件程序执行下述过程：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该处理器还用于调用该存储器中的软件程序执行下述过程：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数后，可以 将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据该内容特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数和关系特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数后，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该处理器还用于调用该存储器中的软件程序执行下述过程：根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该处理器还用于调用该存储器中的软件程序执行下述过程：将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第八方面，本申请提供了一种识别文本的装置，该装置可以包括：至少一个处理器、存储器，该存储器存储有软件程序，该处理器调用该存储器中的软件程序执行下述过程：从该存储器或服务器中获取文本以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的关系信息用于指示该用户与其他用户的层级关系，该服务器与该识别文本的装置连接，该服务器中存储有该文本以及该用户的关系信息；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文 本的接收人和/或该文本的发送人的关系；根据该关系特征系数确定该文本的重要性。
上述第八方面提供的技术方案，可以获取文本以及用户的关系信息，根据文本以及用户的关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据上下级关系特征系数、部门关系特征系数、协 同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该根据该文本以及该用户的关系信息得到关系特征系数，包括：从该存储器或该服务器中获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据该关系特征系数确定该文本的重要性，包括：以该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到关系特征系数后，可以将关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该处理器还用于调用该存储器中的软件程序执行下述过程：将该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将关系特征系数中的系数做乘法运算得到文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该根据该关系特征系数确定该文本的重要性，包括：将该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到关系特征系数后，可以将关系特征系数中的系数做乘法运算得到文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该处理器还用于调用该存储器中的软件程序执行下述过程：根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该处理器还用于调用该存储器中的软件程序执行下述过程：将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第九方面，本申请提供了一种识别文本的装置，该装置可以包括：至少一个处理器、存储器，该存储器存储有软件程序，该处理器用于调用该存储器中的软件程序执行下述过程：从该存储器或服务器中获取文本、用户的画像信息以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示该用户相关的文本中的N个关键词，N为大于或等于1的整数，该用户的关系信息用于指示该用户与其他用户的层级关系，该服务器与该识别文本的装置连接，该服务器中存储有文本、用户的画像信息以及用户的关系信息；根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示文本的内容的重要性；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；根据该内容特征系数以及该关系特征系数确定该文本的重要性。
上述第九方面提供的技术方案，可以获取文本、用户的画像信息以及用户的关系信息，根据文本以及用户的画像信息得到内容特征系数，根据文本以及用户的关系信息得到关系特征系数，并根据内容特征系数和关系特征系数确定该文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该用户的画像信息包括N个标签信息，该N个标签信息对应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。基于上述技术方案，可以根据文本以及N个标签信息得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，或者，可以根据文本、N个标签信息以及N个标签信息中每个标签信息的权重得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，该内容特征系数包括N个系数，其中，第n个系数根据该文本和该用户的画像信息中第n个标签信息得到，n为大于0且小于或等于N的正整数。基于上述技术方案，可以根据文本和用户画像信息中第n个标签信息得到内容特征系数中的第n个系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文本的重要性。
一种可能的实现方式，根据该文本以及该用户的画像信息得到内容特征系数，包括：从该存储器或该服务器中获取第一用户行为反馈数据，其中，第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；从该存储器或该服务器中获取第一对应关系，其中，该第一对应关系用于指示第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；根据该第一用户行为反馈数据和该第一对应关系得到该第一用户行为反馈数据的影响系数；根据该文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到该内容特征系数。基于上述技术方案，可以获取第一用户行为反馈数据，根据第一用户行为反馈数据以及第一关系得到第一用户行为反馈数据的影响系数，根据文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到内容特征系数，并根据内容特征系数和关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或者发送的文 本的重要性。
一种可能的实现方式，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户的上下级关系信息、用户的部门关系信息、协同关系信息以及用户的沟通关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系；该用户的部门关系信息包括该用户与其他用户的最小单元组织；该协同关系信息包括指示信息，该指示信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息包括该用户与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系、用户与其他用户的最小单元组织、指示信息以及该用户与其他用户的沟通频率得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该用户的上下级关系信息包括该用户与其他用户的上下级关系的权重，该用户与其他用户的上下级关系的权重用于指示该用户与其他用户之间的上下级关系；该用户的部门关系信息包括最小单元组织的权重，该最小单元组织的权重用于指示该用户与其他用户的最小单元组织；该协同关系信息包括协同关系权重，该协同关系权重用于指示该用户与其他用户的协同关系；该用户的沟通关系信息包括沟通频率的权重，该沟通频率的权重用于指示用于与其他用户的沟通频率。基于上述技术方案，可以根据文本、用户与其他用户的上下级关系的权重、最小单元组织的权重、协同关系权重以及该沟通频率的权重得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。基于上述技术方案，可以根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数，根据用户的沟通关系信息得到沟通关系特征系数，并根据内容特征系数、上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的 重要性。
一种可能的实现方式，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该根据该文本以及该用户的关系信息得到关系特征系数，包括：从该存储器或该服务器中获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；从该存储器或该服务器中获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。基于上述技术方案，可以获取第二用户行为反馈数据，根据第二用户行为反馈数据以及第二对应关系，得到第二用户行为反馈数据的影响系数，根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容，以及用户与文本的接收人和发送人的关系识别用户接收或发送的文本的重要性。
一种可能的实现方式，该处理器还用于调用该存储器中的软件程序执行下述过程：根据该文本得到该消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数；该确定该文本的重要性，还包括：根据该消息影响范围特征系数确定该文本的重要性。基于上述技术方案，可以根据该文本得到消息影响范围特征系数，根据内容特征系数、关系特征系数以及消息影响范围特征系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及文本的接收人的个数识别用户接收或发送的文本的重要性。
一种可能的实现方式，该处理器还用于调用该存储器中的软件程序执行下述过程：根据该用户的关系信息得到该用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度；该确定该文本的重要性，还包括：根据该用户与接收人/发送人的关系系数确定该文本的重要性。基于上述技术方案，可以根据用户的关系信息得到用户与接收人/发送人的关系系数，根据内容特征系数，关系特征系数以及用户与接收人/发送人的关系系数确定文本的重要性，从而可以根据文本的内容、用户与文本的接收人和发送人的关系以及用户与文本的接收人和/或文本的发送人的关系的紧密程度识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：以该内容特征系数以及该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到该内容特征系数以及该关系特征系数后，可以将该内容特征系数以及该关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学 习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到特征系数集合后，可以将内容特征系数、关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：以该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。基于上述技术方案，在得到特征系数集合后，可以将内容特征系数、关系特征系数和该消息影响范围特征系数和该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过机器学习方法确定该文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该处理器还用于调用该存储器中的软件程序执行下述过程：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该处理器还用于调用该存储器中的软件程序执行下述过程：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，当重要性相同的文本有多个时，该内容特征系数包括N个系数，该处理器还用于调用该存储器中的软件程序执行下述过程：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数对用户接收或发送的文本进行排序，可以使用户根据文本的重要性处理文本。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数和关系特征系数后，可以将该内容特征系数中的系数之和与 该关系特征系数中的系数做乘法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数和该消息影响范围特征系数后，可以将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，该根据内容特征系数以及关系特征系数确定该文本的重要性，包括：将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；根据该文本的重要性系数确定该文本的重要性。基于上述技术方案，在得到该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数后，可以将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数，并根据文本的重要性系数确定文本的重要性，从而可以识别用户接收或发送的文本的重要性。
一种可能的实现方式，当重要性相同的文本有多个时，该处理器还用于调用该存储器中的软件程序执行下述过程：根据该文本的重要性系数对该用户接收或发送的文本进行排序。基于上述技术方案，在确定文本的重要性后，重要性相同的文本有多个时，可以根据文本的重要性系数对用户接收或发送的文本进行排序，以使得用户根据文本的重要性处理文本。
一种可能的实现方式，该处理器还用于调用该存储器中的软件程序执行下述过程：将该文本按照该文本的重要性分类显示。基于上述技术方案，可以将文本按照文本的重要性分类显示，以使得用户对不同重要性类型的文本分别处理。
第十方面，本申请提供了一种系统芯片，该系统芯片可以应用在识别文本的装置中，该系统芯片包括：至少一个处理器，涉及的程序指令在该至少一个处理器中执行，以实现第一方面以及第一方面的任一种可能的实现方式所述的识别文本的方法。可选的，该系统芯片还可以包括至少一个存储器，该存储器存储有涉及的程序指令。
第十一方面，本申请提供了一种系统芯片，该系统芯片可以应用在识别文本的装置中，该系统芯片包括：至少一个处理器，涉及的程序指令在该至少一个处理器中执行，以实现第一方面以及第二方面的任二种可能的实现方式所述的识别文本的方法。可选的，该系统芯片还可以包括至少一个存储器，该存储器存储有涉及的程序指令。
第十二方面，本申请提供了一种系统芯片，该系统芯片可以应用在识别文本的装置中，该系统芯片包括：至少一个处理器，涉及的程序指令在该至少一个处理器中执行，以实现第三方面以及第三方面的任一种可能的实现方式所述的识别文本的方法。可选的，该系统芯片还可以包括至少一个存储器，该存储器存储有涉及的程序指令。
第十三方面，本申请提供了一种计算机存储介质，该计算机存储介质可以应用在识别文本的装置中，该计算机可读存储介质中存储有程序指令，涉及的程序指令运行时，以实现根据第一方面以及第一方面的各种可能的实现方式所述的识别文本的方法。
第十四方面，本申请提供了一种计算机存储介质，该计算机存储介质可以应用在识别文本的装置中，该计算机可读存储介质中存储有程序指令，涉及的程序指令运行时，以实现根据第二方面以及第二方面的各种可能的实现方式所述的识别文本的方法。
第十五方面，本申请提供了一种计算机存储介质，该计算机存储介质可以应用在识别文本的装置中，该计算机可读存储介质中存储有程序指令，涉及的程序指令运行时，以实现根据第三方面以及第三方面的各种可能的实现方式所述的识别文本的方法。
第十六方面，本申请提供了一种计算机程序产品，该计算机程序产品包含程序指令，涉及的程序指令被执行时，以实现根据第一方面以及第一方面的各种可能的实现方式所述的识别文本的方法。
第十七方面，本申请提供了一种计算机程序产品，该计算机程序产品包含程序指令，涉及的程序指令被执行时，以实现根据第二方面以及第二方面的各种可能的实现方式所述的识别文本的方法。
第十八方面，本申请提供了一种计算机程序产品，该计算机程序产品包含程序指令，涉及的程序指令被执行时，以实现根据第三方面以及第三方面的各种可能的实现方式所述的识别文本的方法。
可以理解的，上述提供的任一种系统芯片、计算机存储介质或计算机程序产品等均用于执行上文所提供的对应的方法，因此，其所能达到的有益效果可参考对应的方法中的有益效果，此处不再赘述。
附图说明
图1为本申请实施例提供的显示界面示意图；
图2为本申请实施例提供的硬件设备的硬件结构示意图；
图3为本申请实施例提供的识别文本的方法的流程示意图一；
图4为本申请实施例提供的识别文本的方法的流程示意图二；
图5为本申请实施例提供的识别文本的方法的流程示意图三；
图6为本申请实施例提供的识别文本的方法的流程示意图四；
图7为本申请实施例提供的识别文本的方法的流程示意图五；
图8为本申请实施例提供的识别文本的装置的结构示意图一；
图9为本申请实施例提供的识别文本的装置的结构示意图二；
图10为本申请实施例提供的识别文本的装置的结构示意图三；
图11为本申请实施例提供的识别文本的装置的结构示意图四；
图12为本申请实施例提供的识别文本的装置的结构示意图五；
图13为本申请实施例提供的识别文本的装置的结构示意图六；
图14为本申请实施例提供的识别文本的装置的结构示意图七；
图15为本申请实施例提供的识别文本的装置的结构示意图八；
图16为本申请实施例提供的识别文本的装置的结构示意图九。
具体实施方式
目前，移动办公软件(例如，腾讯通、eSpace和Outlook等软件)已经是人们日常工作中不可或缺的应用。通过移动办公软件，用户每天会接收或发送大量的文本，但是，在这些大量的文本中，仅有部分文本是与用户相关的文本。因此，用户需要从大量的文本中筛选出与用户相关的文本，而如何高效地从大量的文本中筛选出与用户相关的文本是一个急需解决的问题。
为了解决如何从大量的文本中筛选出与用户相关的文本的问题，本申请实施例提供了一种识别文本的方法，该方法可以应用于安装有移动办公软件的装置中，在用户发送或者接收文本后，可以获取文本和用户的画像信息，根据文本和用户的画像信息得到内容特征系数，并根据内容特征系数确定文本的重要性。如此，可以根据文本的内容确定文本的重要性。因此，可以在大量的文本中，自动筛选与用户相关的文本，相对于现有技术中，手动筛选与用户相关的文本，效率更高。
或者，在用户发送或者接收文本后，可以获取文本的接收人的标识和/或文本的发送人的标识，以及用户的关系信息，根据文本的接收人的标识和/或文本的发送人的标识，以及用户的关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性。如此，可以根据用户与文本的接收人和/或文本的发送人的关系确定文本的重要性。因此，可以在大量的文本中，自动筛选与用户相关的文本，相对于现有技术中，手动筛选与用户相关的文本，效率更高。
例如，图1为本申请实施例提供的显示界面示意图，图1中，用户接收到了文本1-文本N，共N个文本。识别文本的装置可以分别获取每个文本和用户的画像信息，根据每个文本和用户的画像信息得到每个文本的内容特征系数，并根据每个文本的内容特征系数分别确定每个文本的重要性，其中，文本2、文本6、文本7以及文本5为重要文本(图1中标记感叹号的文本)，其余文本为非重要文本。
其中，上述名词：文本、用户的画像信息、内容特征系数、文本的接收人的标识(以下简写为接收人的标识)、文本的发送人的标识(以下简写为发送人的标识)、用户的关系信息以及关系特征系数的介绍如下：
文本可以是用户接收或者发送的文本，文本可以包括接收人的标识、发送人的标识以及文本的内容。例如，文本可以是用户通过腾讯通接收的即时消息，或者用户通过邮箱发送的邮件等。本申请下述实施例以文本为用户接收的文本为例进行介绍，文本为用户发送的文本的情况，可以参考本申请下述实施例中文本为用户接收的文本的介绍，不予赘述。
用户的画像信息可以用于指示与用户相关的文本中的N个关键词，N为大于或等于1的整数，例如，用户的画像信息可以用于指示历史上的与用户相关的文本中，出现频率较高的N个词。又例如，用户的画像信息可以用于指示重要项目的名称。
用户的画像信息可以包括N个标签信息，N个标签信息对应指示N个关键词。例如，{tag 1，tag 2，…，tag N}，或者，用户的画像信息可以包括N个标签信息以及N个标签信息中每个标签信息的权重，例如，{tag 1，tag 2，…，tag N}和{tagVal 1，tagVal 2，…，tagVal N}，其中，tagVal 1为tag 1的权重，tagVal 2为tag 2的权重……tagVal N为tag N的权重。其中，标签信息可以包括一个或多个字符。
用户的画像信息可以是用户设置的或者是根据用户历史上发送或接收的文本得到 的。
例如，获取用户历史上的重要本文，从该重要文本中提取出现频率较高的词，将出现频率较高的词作为标签信息，并根据每个标签信息的出现频率设置每个标签信息的权重，标签信息的出现频率越高，该标签信息的权重越大。
内容特征系数可以用于指示文本的内容的重要性。
内容特征系数可以包括N个系数，第n个系数可以根据文本和用户的画像信息中第n个标签信息得到。n为大于0且小于等于N的整数。
接收人的标识可以是接收人的用户名，或者接收人的ID。该接收人的ID可以包括数字和字母。
发送人的标识可以是发送人的用户名，或者发送人的ID。该发送人的ID可以包括数字和字母。
用户的关系信息可以包括用户的组织关系信息以及用户的沟通关系信息。
其中，用户的组织关系信息可以用于指示用户与组织中其他用户(以下简写为用户与其他用户)的层级关系。
可选的，用户的组织关系信息包括用户的上下级关系信息、用户的部门关系信息以及协同关系信息。
下面是用户的上下级关系信息的介绍。
用户的上下级关系信息可以用于指示用户与其他用户的上下级关系。用户与其他用户之间的上下级关系可以包括用户的上级、用户的同级、用户的下级以及不存在上下级关系等等。
一种可能的实现方式，用户的上下级关系信息包括用户与其他用户的上下级关系。
例如，用户1的上下级关系信息可以包括{用户1，用户2，上级}；{用户1，用户3，同级}；{用户1，用户4，下级}；{用户1，用户5，不存在上下级关系}等等，其中，用户2为用户1的上级用户，用户3是用户1同级用户，用户4为用户1的下级用户，用户5为与用户1不存在上下级关系的用户。
另一种可能的实现方式，用户的上下级关系信息包括用户与其他用户的上下级关系的权重。上下级关系的权重可以用于表示用户与其他用户之间的上下级关系。
需要说明的是，用户与其他用户的上下级关系的权重可以包括：用户与上级用户的上下级关系的权重、用户与同级用户的上下级关系的权重、用户与下级用户的上下级关系的权重以及用户与不存在上下级关系的用户的上下级关系的权重。其中，用户与上级用户的上下级关系的权重大于用户与同级用户的上下级关系的权重，用户与同级用户的上下级关系的权重大于用户与下级用户的上下级关系的权重，用户与下级用户的上下级关系的权重大于用户与不存在上下级关系的用户的上下级关系的权重。
示例性的，以用户与上级用户的上下级关系的权重为X，用户与同级用户的上下级关系的权重为Y，用户与下级用户的上下级关系的权重为Z，用户1的上下级关系信息包括{用户1，用户2，X}；{用户1，用户3，Y}；{用户1，用户4，Z}；{用户1，用户5，W}等等为例，则用户2为用户1的上级，X为用户1与用户2的上下级关系的权重，用户3是与用户1同级的用户，Y为用户1与用户3的上下级关系的权重，用户4为用户1的下级，Z为用户1与用户4的上下级关系的权重，用户5为与用户1不存在 上下级关系的用户，W为用户1与用户5的上下级关系的权重，其中，X>Y>Z>W。X、Y、Z以及W为正数。
需要说明的是，用户与其他用户之间的上下级关系除了包括用户的上级、用户的同级以及用户的下级之外，还可以包括其他的上下级关系，例如，用户的上级的上级以及用户的下级的下级等，本申请实施例不予限制。
需要说明的是，用户的上下级关系信息除了上述例子中的形式外，还可以是其他的形式，例如列表形式，本申请实施例不予限制。
下面是用户的部门关系信息的介绍。
用户的部门关系信息可以用于指示用户与其他用户的最小单元组织。用户与其他用户的最小单元组织可以描述为用户与其他用户所在的相同组织中，用户数量最少的组织。
最小单元组织的类型可以包括：小组织、大组织以及第一组织等等，大组织可以包括多个小组织，第一组织可以包括多个大组织。例如，若用户1的最小单元组织为小组织1(即用户1所在的组织中，用户数量最少的组织为小组织1)，用户2的最小单元组织为小组织1，用户3的最小单元组织为小组织2，小组织1和小组织2是大组织1中的组，则用户1和用户2的最小单元组织为小组织1，用户1与用户3的最小单元组织为大组织1，用户2与用户3的最小单元组织也为大组织1。
一种可能的实现方式，用户的部门关系信息包括用户与其他用户的最小单元组织。
例如，用户1的部门关系信息可以包括{用户1，用户2，小组织1}；{用户1，用户3，小组织1}；{用户1，用户4，大组织1}等等，其中，用户1与用户2的最小单元组织为小组织1，用户1与用户3的最小单元组织也为小组织1，用户1与用户4的最小单元组织为大组织1。
另一种可能的实现方式，用户的部门关系信息包括最小单元组织的权重。其中，最小单元组织的权重可以是用户与其他用户的最小单元组织的权重。最小单元组织的权重可以用于指示用户与其他用户的最小单元组织。
需要说明的是，最小单元组织的权重可以包括：小组织的权重、大组织的权重以及第一组织的权重。其中，小组织的权重大于大组织的权重，大组织的权重大于第一组织的权重。
示例性的，以小组织的权重为A，大组织的权重为B，第一组织的权重为C，用户1的部门关系信息包括{用户1，用户2，A}；{用户1，用户3，B}；{用户1，用户4，C}等等为例，则用户2为用户1所在的小组织的用户，A为用户1与用户2的最小单元组织的权重，用户3不是用户1所在的小组织的用户，但是用户3为用户1所在的大组织的用户，B为用户1与用户3的最小单元组织的权重，用户4不是用户1所在的小组织和大组织的用户，但是，用户4是用户1所在的第一组织的用户，C为用户1与用户4的最小单元组织的权重，其中，A>B>C。A、B以及C为正数。
需要说明的是，用户的部门关系信息除了上述例子中的形式外，还可以是其他的形式，例如列表形式，本申请实施例不予限制。
需要说明的是，最小单元组织的类型除了小组织、大组织以及第一组织之外，还可以包括其他类型，本申请实施例不予限制。
下面是协同关系信息的介绍。
协同关系信息可以用于指示用户与其他用户是否协同工作。
一种可能的实现方式，协同关系信息包括指示信息，其中，该指示信息用于指示用户与其他用户是否协同工作。
可选的，该指示信息可以是1比特指示信息。例如，指示信息为0时，可以指示用户与其他用户没有协同工作，指示信息为1时，可以指示用户与其他用户协同工作，反之亦然。
例如，用户1的协同关系信息可以包括{用户1，用户2，0}；{用户1，用户3，1}；{用户1，用户4，1}等等，其中，用户1与用户2的没有协同工作，用户1与用户3协同工作，用户1与用户4协同工作。
另一种可能的实现方式，协同关系信息包括协同关系权重，该协同关系权重可以是用户与其他用户的协同关系权重。用户与其他用户的协同关系权重可以用户指示用户与其他用户的协同关系。
其中一种情况，协同关系权重可以包括：用户与其他用户有协同关系时的协同关系权重，以及用户与其他用户没有协同关系时的协同关系权重。其中，用户与其他用户有协同关系时的协同关系权重，大于用户与其他用户没有协同关系时的协同关系权重。
示例性的，以用户与其他用户有协同关系时的协同关系权重为P，用户与其他用户没有协同关系时的协同关系权重为M为例，用户1的协同关系信息可以包括{用户1，用户2，P}；{用户1，用户3，M}；{用户1，用户4，M}等等，其中，用户1与用户2协同工作，用户1与用户3没有协同工作，用户1与用户4没有协同工作。P>M。P和M为正数。
其中另一种情况，协同关系权重可以包括：用户与其他用户没有协同关系时的协同关系权重、用户与其他用户在重要项目上有协同关系时的协同关系权重、以及用户与其他用户在非重要项目上有协同关系时的协同关系权重。其中，用户与其他用户在重要项目上有协同关系时的协同关系权重，大于用户与其他用户在非重要项目上有协同关系时的协同关系权重，用户与其他用户在非重要项目上有协同关系时的协同关系权重，大于用户与其他用户没有协同关系时的协同关系权重。
示例性的，以用户与其他用户没有协同关系时的协同关系权重为H，用户与其他用户在重要项目上有协同关系时的协同关系权重为I，用户与其他用户在非重要项目上有协同关系时的协同关系权重为J为例，用户1的协同关系信息可以包括{用户1，用户2，H}；{用户1，用户3，I}；{用户1，用户4，J}等等，其中，用户1与用户2没有协同工作，用户1与用户3在重要项目上协同工作，用户1与用户4在非重要项目上协同工作。I>J>H。I、J以及H为正数。
用户的沟通关系信息可以用于指示用户与其他用户的沟通频率。
一种可能的实现方式，用户的沟通关系信息包括沟通频率，其中，该沟通频率为用户与其他用户的沟通频率。例如，该用户1与用户2一天内发送或接收了20条信息，则用户1的沟通关系信息为20。
另一种可能的实现方，用户的沟通关系信息包括沟通频率的权重。其中，沟通频率的权重为用户与其他用户的沟通频率的权重。用户与其他用户的沟通频率的权重用于指 示用户与其他用户的沟通频率。
可选的，沟通频率的权重和用户与其他用户的沟通频率成正相关，即用户与其他用户的沟通频率越高，沟通频率的权重越大，用户与其他用户的沟通频率越低，沟通频率的权重越小。
如表1所示，若用户与其他用户的沟通频率大于等于0次/天，并且小于等于10次/天，沟通频率的权重为R，若用户与其他用户的沟通频率大于等于11次/天，并且小于等于30次/天，沟通频率的权重为S，若用户与其他用户的沟通频率大于等于31次/天，沟通频率的权重为T，其中，T>S>R，T、S以及R为正数。
表1
沟通频率的权重 用户与其他用户的沟通频率(次/天)
R 0-10
S 11-30
T 31及以上
关系特征系数可以用于指示用户与接收人和/或发送人的关系。
关系特征系数可以包括组织关系特征系数以及沟通关系特征系数。组织关系特征系数可以包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数。
若用户为发送人，该上下级关系特征系数可以包括用户与每个接收人的上下级关系特征系数，用户与每个接收人的上下级关系特征系数可以用于指示用户与每个接收人之间的上下级关系。该部门关系特征系数可以包括用户与每个接收人的部门关系特征系数，用户与每个接收人的部门关系特征系数可以用于指示用户与每个接收人的最小单元组织。协同关系特征系数可以包括用户与每个接收人的协同关系特征系数，用户与每个接收人的协同关系特征系数可以用于指示用户与每个接收人是否协同工作。沟通关系特征系数可以包括用户与每个接收人的沟通关系特征系数，用户与每个接收人的沟通关系特征系数可以用于指示用户与每个接收人的沟通频率。
若用户为唯一的接收人，该上下级关系特征系数可以包括用户与发送人的上下级关系特征系数，用户与发送人的上下级关系特征系数可以用于指示用户与发送人之间的上下级关系。该部门关系特征系数可以包括用户与发送人的部门关系特征系数，用户与发送人的部门关系特征系数可以用于指示用户与发送人的最小单元组织。协同关系特征系数可以包括用户与发送人的协同关系特征系数，用户与发送人的协同关系特征系数可以用于指示用户与发送人是否协同工作。沟通关系特征系数可以包括用户与发送人的沟通关系特征系数，用户与发送人的沟通关系特征系数可以用于指示用户与发送人的沟通频率。
若用户为多个接收人中的一个，该上下级关系特征系数可以包括用户与发送人的上下级关系特征系数，以及用户与除用户之外的接收人的上下级关系特征系数，用户与除用户之外的接收人的上下级关系特征系数可以用于指示用户与除用户之外的接收人之间的上下级关系。该部门关系特征系数可以包括用户与发送人的部门关系特征系数，以及用户与除用户之外的接收人的部门关系特征系数，用户与除用户之外的接收人的部门关系特征系数可以用于指示用户与除用户之外的接收人的最小单元组织。协同关系特征系数可以包括用户与发送人的协同关系特征系数，以及用户与除用户之外的接收人的协 同关系特征系数，用户与除用户之外的接收人的协同关系特征系数可以用于指示用户与除用户之外的接收人是否协同工作。沟通关系特征系数可以包括用户与发送人的沟通关系特征系数，以及用户与除用户之外的接收人的沟通关系特征系数，用户与除用户之外的接收人的沟通关系特征系数可以用于指示用户与除用户之外的接收人的沟通频率。
本申请实施例提供的识别文本的装置，可以是一个设备内的一个功能模块。可以理解的是，上述功能既可以是硬件设备中的元件，例如手机中的芯片，也可以是在专用硬件上运行的软件功能，或者是平台(例如，云平台)上实例化的虚拟化功能。
例如，识别文本的装置可以通过图2中的硬件设备200来实现。图2所示为可适用于本申请实施例的硬件设备的硬件结构示意图。该硬件设备200包括至少一个处理器201，通信线路202，存储器203以及至少一个通信接口204。
处理器201可以是一个通用中央处理器(central processing unit，CPU)，微处理器，特定应用集成电路(application-specific integrated circuit，ASIC)，或一个或多个用于控制本申请方案程序执行的集成电路。
通信线路202可包括一通路，在上述组件之间传送信息，例如总线。
通信接口204，使用任何收发器一类的装置，用于与其他设备或通信网络通信，如以太网接口，无线接入网接口(radio access network，RAN)，无线局域网接口(wireless local area networks，WLAN)等。
存储器203可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(electrically erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过通信线路202与处理器相连接。存储器也可以和处理器集成在一起。本申请实施例提供的存储器通常可以具有非易失性。其中，存储器203用于存储执行本申请方案所涉及的计算机执行指令，并由处理器201来控制执行。处理器201用于执行存储器203中存储的计算机执行指令，从而实现本申请实施例提供的方法。
可选的，本申请实施例中的计算机执行指令也可以称之为应用程序代码，本申请实施例对此不做具体限定。
在具体实现中，作为一种实施例，处理器201可以包括一个或多个CPU，例如图2中的CPU0和CPU1。
在具体实现中，作为一种实施例，硬件设备200可以包括多个处理器，例如图2中的处理器201和处理器207。这些处理器中的每一个可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
在具体实现中，作为一种实施例，硬件设备200还可以包括输出设备205和输入设备206。输出设备205和处理器201通信，可以以多种方式来显示信息。例如，输出设 备205可以是液晶显示器(liquid crystal display，LCD)，发光二级管(light emitting diode，LED)显示设备，阴极射线管(cathode ray tube，CRT)显示设备，或投影仪(projector)等。输入设备206和处理器201通信，可以以多种方式接收用户的输入。例如，输入设备206可以是鼠标、键盘、触摸屏设备或传感设备等。
上述的硬件设备200可以是一个通用设备或者是一个专用设备。在具体实现中，硬件设备200可以是台式机、便携式电脑、掌上电脑(personal digital assistant，PDA)、移动手机、平板电脑、无线终端设备、嵌入式设备或有图2中类似结构的设备。本申请实施例不限定硬件设备200的类型。
下面以文本为用户接收的文本为例，对本申请实施例提供的识别文本的方法进行具体阐述。
需要说明的是，本申请下述实施例中的信息的名字或参数的名字等只是一个示例，具体实现中也可以是其他的名字，本申请实施例对此不做具体限定。
可以理解的，本申请实施例中，识别文本的装置可以执行本申请实施例中的部分或全部步骤，这些步骤仅是示例，本申请实施例还可以执行其它步骤或者各种步骤的变形。此外，各个步骤可以按照本申请实施例呈现的不同的顺序来执行，并且有可能并非要执行本申请实施例中的全部步骤。
实施例1：
用户接收文本后，识别文本的装置可以获取文本和用户的画像信息，根据文本和用户的画像信息得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容确定文本的重要性。如图3所示，该识别文本的方法可以包括步骤301-步骤303。
步骤301：获取文本和用户的画像信息。
其中，文本以及用户的画像信息的描述可以参考上述名词介绍，此处不予赘述。
可选的，获取文本和用户的画像信息，包括：从远端(例如，服务器)和/或本地获取文本和用户的画像信息。
该服务器可以与识别文本的装置连接，该服务器中存储有文本和/或用户的画像信息。
下面结合图2介绍获取文本和用户的画像信息的具体过程。
例如，若文本与用户的画像信息存储于存储器203中，处理器201调用接口从存储器203中获取文本和用户的画像信息。
又例如，若文本存储于存储器203中，用户的画像信息存储于服务器中，处理器201调用接口从存储器203中获取文本，处理器201通过通信接口204向服务发送请求，该请求包括用户的标识，服务器接收到该请求后，发送用户的画像信息，处理器201通过通信接口204接收该用户的画像信息。
需要说明的是，步骤301中可以用文本的内容替换文本。后续，可以根据文本的内容和用户的画像信息，得到内容特征系数。
步骤302：根据文本和用户的画像信息，得到内容特征系数。
其中，内容特征系数的描述可以参考上述名词解释，此处不予赘述。
可选的，内容特征系数与文本的重要性成正相关，即在其他特征系数不变的情况下， 内容特征系数越大，文本的重要性越高。
步骤302的具体过程可以参考以下示例：
示例1：若用户的画像信息包括N个标签信息，可以根据用户的画像信息中标签信息在文本中出现的次数，确定内容特征系数。
例如，若用户的画像信息中第n个标签信息在文本中出现，则第n个标签信息对应的系数可以为1；若用户的画像信息中第n个标签信息未在文本中出现，则第n个标签信息对应的系数可以为0。
又例如，可以根据表2所示的对应关系确定内容特征系数。表2中，若用户的画像信息中第n个标签信息在文本中未出现，则第n个标签信息对应的系数可以为0；若用户的画像信息中第n个标签信息在文本中出现的次数大于等于1次且小于等于8次，则第n个标签信息对应的系数可以为1；若用户的画像信息中第n个标签信息在文本中出现的次数大于等于9次，则第n个标签信息对应的系数可以为2。
表2
标签信息在文本中出现的次数 该标签信息对应的系数
0次 0
1-8次 1
9次及以上 2
示例2：若用户的画像信息包括N个标签信息以及N个标签信息中每个标签信息的权重，可以根据用户的画像信息中标签信息在文本中出现的次数以及该标签信息的权重，确定内容特征系数。例如，可以根据用户的画像信息中第n个标签信息在文本中出现的次数，得到第n个标签信息的第一参数，将第n个标签信息的第一参数以及权重的乘积作为第n个标签信息对应的系数。
示例性的，以用户的画像信息包括{tag 1，tag 2，tag 3}和{tagVal 1，tagVal 2，tagVal 3}，tag 1和tag 2未在文本中出现，tag 3在文本中出现，若标签信息在文本中出现，则该标签信息的第一参数为1，若标签信息未在文本中出现，则该标签信息的第一参数为0为例，tag 1和tag 2的第一参数为0，tag 3的第一参数为1，tag 1和tag 2对应的系数为0，tag 3对应的系数为tagVal 3，因此，内容特征系数为{0，0，tagVal 3}。
示例3：若用户的画像信息包括N个标签信息，可以从文本中提取文本的关键词，获取文本的关键词与每个标签信息的相似度，将该相似度作为内容特征系数。
可选的，文本的关键词可以包括从文本中提取的单词和/或词组。例如，使用分词软件(例如，jieba软件、ansj软件或者HanLP软件等)从文本中提取的单词和/或词组。文本的关键词可以包括一个或多个单词和/或词组。
进一步可选的，文本的关键词包括标签信息或者标签信息的同义词(或近义词)。
文本的关键词与标签信息的相似度可以用于指示文本的关键词与标签信息的相似程度。文本的关键词与标签信息的相似度可以包括相同、不相同两种，也可以包括相同、不相同、相近三种。
示例性的，可以比较文本的关键词与标签信息，若二者包括的字符完全相同，则确定文本的关键词与标签信息相同；若二者包括的字符中不相同的字符的个数大于或等于第一预设阈值，则确定文本的关键词与标签信息不相同；若二者包括的字符中存在不同 的字符，且不相同的字符的个数小于或等于第二预设阈值，则确定文本的关键词与标签信息相近。
示例性的，本申请实施例中，可以用0、1、2来表示文本的关键词与标签信息的相似度。例如，若文本的关键词与标签信息相同，则文本的关键词与该标签信息的相似度为1，该标签信息对应的系数为1；若文本的关键词与标签信息不相同，则文本的关键词与该标签信息的相似度为0，该标签信息对应的系数为0。本领域技术人员应该理解，文本的关键词与标签信息的相似度还可以通过其他方式或者字符来表示，不予限制。
示例4：若用户的画像信息包括N个标签信息以及N个标签信息中每个标签信息的权重，可以从文本中提取文本的关键词，获取文本的关键词与每个标签信息的相似度，将第n个相似度与该相似度对应的标签信息的权重做乘法运算，得到第n个第二参数，将第n个第二参数作为内容特征系数中第n个系数。
示例性的，以用户的画像信息包括{tag 1，tag 2，…，tag N}和{tagVal 1，tagVal 2，…，tagVal N}，文本的关键词与用户的画像信息中每个标签信息的相似度为{similarCoff 1，similarCoff 2，…，similarCoff N}(similarCoff 1为文本的关键词与tag 1的相似度，similarCoff 2为文本的关键词与tag 2的相似度，以此类推)为例，内容特征系数可以为{contentCoff 1，contentCoff 2，…，contentCoff N}，其中，contentCoff n＝similarCoff n*tagVal n。
进一步可选的，将第n个第二参数作为第一函数的参数，该第一函数的值作为内容特征系数中第n个系数，其中，该函数的导数大于且等于0。
示例性的，以用户的画像信息包括{tag 1，tag 2，…，tag N}和{tagVal 1，tagVal 2，…，tagVal N}，文本的关键词与用户的画像信息中每个标签信息的相似度为{similarCoff 1，similarCoff 2，…，similarCoff N}(similarCoff 1为文本的关键词与tag 1的相似度，similarCoff 2为文本的关键词与tag 2的相似度，以此类推)为例，内容特征系数可以为{contentCoff 1，contentCoff 2，…，contentCoff N}，其中，contentCoff n＝Function(x)，Function(x)为x的函数，Function′(x)≥0，x＝similarCoff n*tagVal n。
可选的，内容特征系数除了根据文本以及用户的画像信息得到，还可以根据文本、用户的画像信息以及第一用户行为反馈数据得到。
其中，第一用户行为反馈数据可以用于指示历史上与该文本内容相似度高的文本的重要性。第一用户行为反馈数据可以在步骤301中获取，也可以在步骤302之前获取。后续，可以根据第一用户行为反馈数据得到第一用户行为反馈数据的影响系数，在根据上述示例1-示例4所述的方法得到内容特征系数后，可以通过第一用户行为反馈数据的影响系数对内容特征系数进行修正，使得修正后的内容特征系数更准确。具体的，可以参考下述示例5中的描述。
示例5：可以根据第一用户行为反馈数据得到第一用户行为反馈数据的影响系数，将上述示例1-示例4所述的方法得到内容特征系数与第一用户行为反馈数据的影响系数做加减运算，得到更准确的内容特征系数。
其中，第一用户行为反馈数据的影响系数可以用于指示历史上与该文本内容相似度高的文本的重要性。
可选的，第一用户行为反馈数据的影响系数与文本的重要性成正相关，即在其他特征系数不变的情况下，第一用户行为反馈数据的影响系数越大，文本的重要性越高。
可选的，根据第一用户行为反馈数据得到第一用户行为反馈数据的影响系数，包括：获取第一对应关系，其中，该第一对应关系用于指示第一用户行为反馈数据的影响系数与第一用户行为反馈数据之间的关联关系；根据第一用户行为反馈数据，以及第一对应关系，确定第一用户行为反馈数据的影响系数。
其中，第一对应关系可以是预设置的，并存储在图2中的存储器203或服务器中。
可选的，获取第一对应关系，包括从图2中的存储器203或服务器中获取第一对应关系。
如表3所示，为第一用户行为反馈数据的影响系数与第一用户行为反馈数据的对应关系。表3中，若第一用户行为反馈数据指示历史上与该文本内容相似度高的文本为重要文本，则第一用户行为反馈数据的影响系数为a，若第一用户行为反馈数据指示历史上与该文本内容相似度高的文本为非重要文本，则第一用户行为反馈数据的影响系数为-a。
表3
Figure PCTCN2020095510-appb-000001
示例性的，以通过上述示例1-示例4所述的方法得到内容特征系数为{contentCoff 1，contentCoff 2，…，contentCoff N}，第一用户行为反馈数据的影响系数与第一用户行为反馈数据的对应关系如表3所示为例，若第一用户行为反馈数据指示历史上与该文本内容相似度高的文本为重要文本，则第一用户行为反馈数据的影响系数为a，考虑了第一用户行为反馈数据的影响系数后的新的内容特征系数为{contentCoff 1+a，contentCoff 2+a，…，contentCoff N+a}。
步骤303：根据内容特征系数确定文本的重要性。
一种可能的实现方式，根据内容特征系数确定文本的重要性，包括：以内容特征系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性。
其中，文本的重要性可以有多种形式，例如，重要文本和非重要文本。又例如，一级重要文本、二级重要文本、三级重要文本以及非重要文本，其中，一级重要文本的重要性大于二级重要文本的重要性，二级重要文本的重要性大于三级重要文本的重要性，三级重要文本的重要性大于非重要文本的重要性。
其中，机器学习方法可以包括传统的机器学习方法以及深度学习方法。
传统的机器学习方法可以包括：线性判别分析(linear discriminant analysis，LDA)、支持向量机(support vector machine，SVM)以及增强学习方法(adboost)等等。
深度学习方法可以包括：深度神经网络(deep neural networks，DNN)、卷积神经网络(convolutional neural networks，CNN)以及循环神经网络(recurrent neural networks，RNN)等等。
可选的，当重要性相同的文本有多个时(例如，一级重要文本有多个时)，根据内容特征系数确定文本的重要性之后还包括：将内容特征系数中的N个系数全部或部分做 加法运算得到文本的重要性系数；根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，将内容特征系数中的N个系数全部或部分做加法运算得到文本的重要性系数，包括：根据公式textImportantCoff＝contentCoff 1+contentCoff 2+…+contentCoff N计算文本的重要性系数。其中，textImportantCoff为文本的重要性系数，contentCoff 1、contentCoff 2、…、contentCoff N为内容特征系数。
另一种可能的实现方式，根据内容特征系数确定文本的重要性，包括：将内容特征系数中的N个系数全部或部分做加法运算得到文本的重要性系数；根据文本的重要性系数确定文本的重要性。
将内容特征系数中的N个系数全部或部分做加法运算得到文本的重要性系数可以包括：根据公式textImportantCoff＝contentCoff 1+contentCoff 2+…+contentCoff N计算文本的重要性系数。
可选的，根据文本的重要性系数确定文本的重要性，包括：若文本的重要性系数大于或等于第一阈值，则确定该文本为重要文本；或者，根据文本的重要性系数与文本的重要性的对应关系，确定文本的重要性。
如表4所示，为文本的重要性系数与文本的重要性的对应关系。表4中，若文本的重要性系数大于或等于9，则文本的重要性为一级重要文本，若文本的重要性系数为6-8，则文本的重要性为二级重要文本，若文本的重要性系数为2-5，则文本的重要性为三级重要文本，若文本的重要性系数维0-1，则文本的重要性为非重要文本。
表4
文本的重要性系数 文本的重要性
大于或等于9 一级重要文本
6-8 二级重要文本
2-5 三级重要文本
0-1 非重要文本
需要说明的是，表4仅是文本的重要性系数与文本的重要性的对应关系的示例，在实际应用中，文本的重要性系数与文本的重要性的对应关系还可以是其他形式，而且文本的重要性系数与文本的重要性的对应关系可以为上述表1中的某一行、某些行、表1中的全部、或者比表1示出的更多的对应关系，本申请不进行具体限制。
进一步可选的，当重要性相同的文本有多个时，根据内容特征系数确定文本的重要性之后还包括：根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，在根据内容特征系数确定文本的重要性之后，将文本按照文本的重要性分类显示。例如，图2中的处理器201可以通过调用接口将文本按照文本的重要性分类显示在输出设备205中(例如，将重要文本显示在非重要文本的前面)。
可选的，若用户更换设备，可以通过以下两种方法将旧设备中的文本的重要性传输到新设备中：
方法1：通过显示指令，从旧设备的数据存储区中提取文本的标识、文本的重要性和/或文本的重要性系数，进行加密，将加密后的文本的标识、文本的重要性和/或文本的重要性系数通过网络安全传输方式传输到服务器，新设备可以通过显示指令从服务器 上下载加密后的文本的标识、文本的重要性和/或文本的重要性系数。
方式2：通过直接拷贝的方式将旧设备中的文本的标识、文本的重要性和/或文本的重要性系数拷贝到新设备中。
基于图3所示的方法，可以获取文本和用户的画像信息，根据文本和用户的画像信息得到内容特征系数，并根据内容特征系数确定文本的重要性，从而可以根据文本的内容确定文本的重要性。
实施例2：
用户接收文本后，可以获取接收人的标识和/或发送人的标识，以及用户的关系信息，根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与接收人和/或发送人的关系确定文本的重要性。如图4所示，该识别文本的方法可以包括步骤401-步骤403。
步骤401：获取接收人的标识和/或发送人的标识，以及用户的关系信息。
其中，接收人的标识和/或发送人的标识，以及用户的关系信息的描述可以参考上述名词介绍，此处不予赘述。
可选的，获取接收人的标识和/或发送人的标识，以及用户的关系信息，包括：从远端(例如：服务器)和/或从本地获取接收人的标识和/或发送人的标识，以及用户的关系信息。
该服务器可以与该识别文本的装置连接，该服务器中存储有接收人的标识和/或发送人的标识，和/或用户的关系信息。
下面结合图2介绍获取接收人的标识和/或发送人的标识，以及用户的关系信息。
例如，若文本与用户的关系信息存储于存储器203中，处理器201调用接口从存储器203中获取接收人的标识和/或发送人的标识，以及用户的关系信息。
又例如，若文本存储于存储器203中，用户的关系信息存储于服务器中，处理器201调用接口从存储器203中获取接收人的标识和/或发送人的标识，处理器201通过通信接口204向服务发送请求，该请求包括用户的标识，服务器接收到该请求后，发送用户的关系信息，处理器201通过通信接口204接收该用户的关系信息。
可选的，若用户为发送人，获取接收人的标识以及用户的关系信息；若用户为文本的唯一接收人，获取发送人的标识以及用户的关系信息；若用户为多个接收人中的一个，获取接收人的标识、发送人的标识，以及用户的关系信息。
需要说明的是，未有人员变动的时候，用户的关系信息一般是固定不变的，因此，可以在第一次发送或接收文本时，获取该用户的关系信息，后续，在接收或发送新文本时，不需要再次获取用户的关系信息，仅获取接收人的标识和/或发送人的标识。后续，当有人员变动的时候(例如，有人升职或离职等等)，管理员将该变动更新到本地或服务器后，可以向识别文本的装置发送通知消息，该识别文本的装置接收到该通知消息后再次获取用户的关系信息。
步骤402：根据接收人的标识和/或发送人的标识，以及用户的关系信息，得到关系特征系数。
可选的，关系特征系数与文本的重要性成正相关，即在其他特征系数不变的情况下，关系特征系数越大，文本的重要性越高。
用户的关系信息可以包括用户的组织关系信息以及用户的沟通关系信息。其中，用户的组织关系信息包括用户的上下级关系信息、用户的部门关系信息以及协同关系信息。用户的上下级关系信息、用户的部门关系信息、协同关系信息用户的沟通关系信息的描述可以参考上述名词介绍，不予赘述。
关系特征系数包括：组织关系特征系数以及沟通关系特征系数。其中，组织关系特征系数可以包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数。上下级关系特征系数、部门关系特征系数、协同关系特征系数以及沟通关系特征系数的描述可以参考上述名词介绍，不予赘述。
其中，组织关系特征系数可以根据用户的组织关系信息得到。沟通关系特征系数可以根据用户的沟通关系信息得到。
可选的，组织关系特征系数可以根据用户的组织关系信息得到，包括：根据用户的上下级关系信息得到上下级关系特征系数，根据用户的部门关系信息得到部门关系特征系数，根据协同关系信息得到协同关系特征系数。
根据用户的上下级关系信息得到上下级关系特征系数可以有以下4种情况：
情况1：用户为唯一的接收人，用户的上下级关系信息包括用户与其他用户的上下级关系。
若用户为唯一的接收人，上下级关系特征系数可以包括用户与发送人的上下级关系特征系数。根据用户的上下级关系信息得到上下级关系特征系数，包括：从用户与其他用户的上下级关系中获取用户与发送人的上下级关系，根据用户与发送人的上下级关系，得到用户与发送人的上下级关系特征系数。
一种可能的实现方式，根据用户与发送人的上下级关系，得到用户与发送人的上下级关系特征系数，包括：根据用户与发送人的上下级关系，按照上下级关系特征系数的设置规则确定用户与发送人的上下级关系特征系数。
其中，上下级关系特征系数的设置规则可以如下所示：用户与上级用户的上下级关系特征系数大于用户与同级用户的上下级关系特征系数，用户与同级用户的上下级关系特征系数大于用户与下级用户的上下级关系特征系数，用户与下级用户的上下级关系特征系数大于用户与不存在上下级关系的用户的上下级关系特征系数。
示例性的，以用户1的上下级关系信息包括{用户1，用户2，上级}；{用户1，用户3，同级}；{用户1，用户4，下级}；{用户1，用户5，不存在上下级关系}，用户与用户的上级用户的上下级关系特征系数为6，用户与用户的同级用户的上下级关系特征系数为4，用户与用户的下级用户的上下级关系特征系数为2，用户不存在上下级关系的用户与用户的上下级关系特征系数为0.01为例，若发送人的标识为用户2，则用户与发送人的上下级关系特征系数可以是6；若发送人的标识为用户3，则用户与发送人的上下级关系特征系数可以是4；若发送人的标识为用户4，则用户与发送人的上下级关系特征系数可以是2；若发送人的标识为用户5，则用户与发送人的上下级关系特征系数可以是0.01。
情况2：用户为唯一的接收人，用户的上下级关系信息可以包括用户与其他用户的上下级关系的权重。
若用户为唯一的接收人，上下级关系特征系数可以包括用户与发送人的上下级关系 特征系数。根据用户的上下级关系信息得到上下级关系特征系数，包括：从用户与其他用户的上下级关系的权重中获取用户与发送人的上下级关系的权重，根据用户与发送人的上下级关系的权重，得到用户与发送人的上下级关系特征系数。
一种可能的实现方式，根据用户与发送人的上下级关系的权重，得到用户与发送人的上下级关系特征系数，包括：将用户与发送人的上下级关系的权重作为用户与发送人的上下级关系特征系数。
示例性的，以用户1的上下级关系信息包括{用户1，用户2，X}；{用户1，用户3，Y}；{用户1，用户4，Z}；{用户1，用户5，W}，其中，X为用户1与用户2的上下级关系的权重，Y为用户1与用户3的上下级关系的权重，Z为用户1与用户4的上下级关系的权重，W为用户1与用户5的上下级关系的权重为例，若发送人的标识为用户2，则用户与发送人的上下级关系特征系数可以是X；若发送人的标识为用户3，则用户与发送人的上下级关系特征系数可以是Y；若发送人的标识为用户4，则用户与发送人的上下级关系特征系数可以是Z；若发送人的标识为用户5，则用户与发送人的上下级关系特征系数可以是W。
需要说明的是，除了将用户与发送人的上下级关系的权重作为用户与发送人的上下级关系特征系数之外，还可以将用户与发送人的上下级关系的权重进行处理(例如，加一个数或者减一个数等)，并将处理后的数据作为用户与发送人的上下级关系特征系数。其中，处理后的数据应满足上述上下级关系特征系数的设置规则。
情况3：用户为多个接收人中的一个，用户的上下级关系信息可以包括用户与其他用户的上下级关系。
若用户为多个接收人中的一个，上下级关系特征系数包括用户与发送人的上下级关系特征系数，以及用户与除用户之外的接收人的上下级关系特征系数。根据用户的上下级关系信息得到上下级关系特征系数，包括：从用户与其他用户的上下级关系中获取用户与发送人的上下级关系，以及用户与除用户之外的接收人的上下级关系，根据用户与发送人的上下级关系，得到用户与发送人的上下级关系特征系数，根据用户与除用户之外的接收人的上下级关系，得到用户与除用户之外的接收人的上下级关系特征系数。
其中，根据用户与发送人的上下级关系，得到用户与发送人的上下级关系特征系数的具体描述可以参考上述情况1中的介绍，不予赘述。
其中，根据用户与除用户之外的接收人的上下级关系，得到用户与除用户之外的接收人的上下级关系特征系数的具体描述可以参考上述情况1中，对根据用户与发送人的上下级关系，得到用户与发送人的上下级关系特征系数的介绍，不予赘述。
情况4：用户为多个接收人中的一个，用户的上下级关系信息可以包括用户与其他用户的上下级关系的权重。
若用户为多个接收人中的一个，上下级关系特征系数包括用户与发送人的上下级关系特征系数，以及用户与除用户之外的接收人的上下级关系特征系数。根据用户的上下级关系信息得到上下级关系特征系数，包括：从用户与其他用户的上下级关系的权重中获取用户与发送人的上下级关系的权重，以及用户与除用户之外的接收人的上下级关系的权重，根据用户与发送人的上下级关系的权重，得到用户与发送人的上下级关系特征系数，根据用户与除用户之外的接收人的上下级关系的权重，得到用户与除用户之外的 接收人的上下级关系特征系数。
其中，根据用户与发送人的上下级关系的权重，得到用户与发送人的上下级关系特征系数的具体描述可以参考上述情况2中的介绍，不予赘述。
其中，根据用户与除用户之外的接收人的上下级关系的权重，得到用户与除用户之外的接收人的上下级关系特征系数的具体描述可以参考上述情况2中，对根据用户与发送人的上下级关系的权重，得到用户与发送人的上下级关系特征系数的介绍，不予赘述。
根据用户的部门关系信息得到部门关系特征系数可以有以下4种情况：
情况5：用户为唯一的接收人，用户的部门关系信息包括用户与其他用户的最小单元组织。
若用户为唯一的接收人，部门关系特征系数包括用户与发送人的部门关系特征系数。根据用户的部门关系信息得到部门关系特征系数，包括：从用户与其他用户的最小单元组织中获取用户与发送人的最小单元组织，根据用户与发送人的最小单元组织，得到用户与发送人的部门关系特征系数。
一种可能的实现方式，根据用户与发送人的最小单元组织，得到用户与发送人的部门关系特征系数，包括：根据用户与发送人的最小单元组织，按照部门关系特征系数的设置规则确定用户与发送人的部门关系特征系数。
其中，部门关系特征系数的设置规则可以如下所示：小组织中的用户的部门关系特征系数大于大组织中的用户的部门关系特征系数，大组织中的用户的部门关系特征系数大于第一组织中的用户的部门关系特征系数。
示例性的，以用户1的部门关系信息包括{用户1，用户2，小组织1}；{用户1，用户3，小组织1}；{用户1，用户4，大组织1}，小组织中的用户的部门关系特征系数为5，大组织中的用户的部门关系特征系数为2，第一组织中的用户的部门关系特征系数为0.01为例，若发送人的标识为用户2，则用户与发送人的部门关系特征系数可以是5；若发送人的标识为用户3，则用户与发送人的部门关系特征系数可以是2；若发送人的标识为用户4，则用户与发送人的部门关系特征系数可以是0.01。
情况6：用户为唯一的接收人，用户的部门关系信息包括用户与其他用户的最小单元组织的权重。
若用户为唯一的接收人，部门关系特征系数包括用户与发送人的部门关系特征系数。根据用户的部门关系信息得到部门关系特征系数，包括：从用户与其他用户的最小单元组织的权重中获取用户与发送人的最小单元组织的权重，根据用户与发送人的最小单元组织的权重，得到用户与发送人的部门关系特征系数。
一种可能的实现方式，根据用户与发送人的最小单元组织的权重，得到用户与发送人的部门关系特征系数，包括：将用户与发送人的最小单元组织的权重作为用户与发送人的部门关系特征系数。
示例性的，以用户1的部门关系信息包括{用户1，用户2，A}；{用户1，用户3，B}；{用户1，用户4，C}，其中，A为用户1与用户2的最小单元组织的权重，B为用户1与用户3的最小单元组织的权重，C为用户1与用户4的最小单元组织的权重为例，若发送人的标识为用户2，则用户与发送人的部门关系特征系数可以是A；若发送人的标识为用户3，则用户与发送人的部门关系特征系数可以是B；若发送人的标识为用户 4，则用户与发送人的部门关系特征系数可以是C。
需要说明的是，除了将用户与发送人的最小单元组织的权重作为用户与发送人的部门关系特征系数之外，还可以将用户与发送人的最小单元组织的权重进行处理(例如，加一个数或者减一个数等)，并将处理后的数据作为用户与发送人的部门关系特征系数。其中，处理后的数据应满足上述部门关系特征系数的设置规则。
情况7：用户为多个接收人中的一个，用户的部门关系信息包括用户与其他用户的最小单元组织。
若用户为多个接收人中的一个，部门关系特征系数包括用户与发送人的部门关系特征系数，以及用户与除用户之外的接收人的部门关系特征系数。根据用户的部门关系信息得到部门关系特征系数，包括：从用户与其他用户的最小单元组织中获取用户与发送人的最小单元组织，以及用户与除用户之外的接收人的最小单元组织，根据用户与发送人的最小单元组织，得到用户与发送人的部门关系特征系数，根据用户与除用户之外的接收人的最小单元组织，得到用户与除用户之外的接收人的部门关系特征系数。
其中，根据用户与发送人的最小单元组织，得到用户与发送人的部门关系特征系数的具体描述可以参考上述情况5中的介绍，不予赘述。
其中，根据用户与除用户之外的接收人的最小单元组织，得到用户与除用户之外的接收人的部门关系特征系数的具体描述可以参考上述情况5中，对根据用户与发送人的最小单元组织，得到用户与发送人的部门关系特征系数的介绍，不予赘述。
情况8：用户为多个接收人中的一个，用户的部门关系信息包括用户与其他用户的最小单元组织的权重。
若用户为多个接收人中的一个，部门关系特征系数包括用户与发送人的部门关系特征系数，以及用户与除用户之外的接收人的部门关系特征系数。根据用户的部门关系信息得到部门关系特征系数，包括：从用户与其他用户的最小单元组织的权重中获取用户与发送人的最小单元组织的权重，以及用户与除用户之外的接收人的最小单元组织的权重，根据用户与发送人的最小单元组织的权重，得到用户与发送人的部门关系特征系数，根据用户与除用户之外的接收人的最小单元组织的权重，得到用户与除用户之外的接收人的部门关系特征系数。
其中，根据用户与发送人的最小单元组织的权重，得到用户与发送人的部门关系特征系数的具体描述可以参考上述情况6中的介绍，不予赘述。
其中，根据用户与除用户之外的接收人的最小单元组织的权重，得到用户与除用户之外的接收人的部门关系特征系数的具体描述可以参考上述情况6中，对根据用户与发送人的最小单元组织的权重，得到用户与发送人的部门关系特征系数的介绍，不予赘述。
根据协同关系信息得到协同关系特征系数可以有以下4种情况：
情况9：用户为唯一的接收人，协同关系信息包括指示用户与其他用户是否有协同工作的指示信息。
若用户为唯一的接收人，协同关系特征系数包括用户与发送人的协同关系特征系数。根据协同关系信息得到协同关系特征系数，包括：从指示用户与其他用户是否有协同工作的指示信息中获取指示用户与发送人是否有协同工作的指示信息，根据指示用户与发送人是否有协同工作的指示信息，得到用户与发送人的协同关系特征系数。
一种可能的实现方式，根据指示用户与发送人是否有协同工作的指示信息，得到用户与发送人的协同关系特征系数，包括：根据指示用户与发送人是否有协同工作的指示信息，按照协同关系特征系数的设置规则确定用户与发送人的协同关系特征系数。
其中，若协同关系特征系数包括：用户与其他用户有协同关系时的协同关系特征系数，以及用户与其他用户没有协同关系时的协同关系特征系数，协同关系特征系数的设置规则可以如下所示：用户与其他用户有协同关系时的协同关系特征系数，大于用户与其他用户没有协同关系时的协同关系特征系数。若协同关系特征系数包括：用户与其他用户没有协同关系时的协同关系特征系数、用户与其他用户在重要项目上有协同关系时的协同关系特征系数、以及用户与其他用户在非重要项目上有协同关系时的协同关系特征系数，协同关系特征系数的设置规则可以如下所示：用户与其他用户在重要项目上有协同关系时的协同关系特征系数，大于用户与其他用户在非重要项目上有协同关系时的协同关系特征系数，用户与其他用户在非重要项目上有协同关系时的协同关系特征系数，大于用户与其他用户没有协同关系时的协同关系特征系数。
示例性的，以用户1的协同关系信息包括{用户1，用户2，0}；{用户1，用户3，1}；{用户1，用户4，1}，0指示用户与其他用户没有协同工作，1指示用户与其他用户协同工作，用户与其他用户有协同关系时的协同关系特征系数为4，用户与其他用户没有协同关系时的协同关系特征系数为0.01为例，若发送人的标识为用户2，则用户与发送人的协同关系特征系数可以是0.01；若发送人的标识为用户3，则用户与发送人的协同关系特征系数可以是4；若发送人的标识为用户4，则用户与发送人的协同关系特征系数可以是4。
情况10：用户为唯一的接收人，协同关系信息包括用户与其他用户的协同关系权重。
若用户为唯一的接收人，协同关系特征系数包括用户与发送人的协同关系特征系数。根据协同关系信息得到协同关系特征系数，包括：从用户与其他用户的协同关系权重中获取指示用户与发送人的协同关系权重，根据用户与发送人的协同关系权重，得到用户与发送人的协同关系特征系数。
一种可能的实现方式，根据用户与发送人的协同关系权重，得到用户与发送人的协同关系特征系数，包括：将用户与发送人的协同关系权重作为用户与发送人的协同关系特征系数。
示例性的，用户1的协同关系信息可以包括{用户1，用户2，P}；{用户1，用户3，M}，其中，P为用户1与用户2的协同关系权重，M为用户1与用户3的协同关系权重为例，若发送人的标识为用户2，则用户与发送人的协同关系特征系数可以是P；若发送人的标识为用户3，则用户与发送人的协同关系特征系数可以是M。
需要说明的是，除了将用户与发送人的协同关系权重作为用户与发送人的协同关系特征系数之外，还可以将用户与发送人协同关系权重进行处理(例如，加一个数或者减一个数等)，并将处理后的数据作为用户与发送人的协同关系特征系数。其中，处理后的数据应满足上述协同关系特征系数的设置规则。
情况11：用户为多个接收人中的一个，协同关系信息包括指示用户与其他用户是否有协同工作的指示信息。
若用户为多个接收人中的一个，协同关系特征系数包括用户与发送人的协同关系特 征系数，以及用户与除用户之外的接收人的协同关系特征系数。根据用户的协同关系信息得到协同关系特征系数，包括：从指示用户与其他用户是否有协同工作的指示信息中获取指示用户与发送人是否有协同工作的指示信息，以及指示用户与除用户之外的接收人是否有协同工作的指示信息，根据指示用户与发送人是否有协同工作的指示信息，得到用户与发送人的协同关系特征系数，根据指示用户与除用户之外的接收人是否有协同工作的指示信息，得到用户与除用户之外的接收人的协同关系特征系数。
其中，根据指示用户与发送人是否有协同工作的指示信息，得到用户与发送人的协同关系特征系数的具体描述可以参考上述情况9中的介绍，不予赘述。
其中，根据指示用户与除用户之外的接收人是否有协同工作的指示信息，得到用户与除用户之外的接收人的协同关系特征系数的具体描述可以参考上述情况9中，对根据指示用户与发送人是否有协同工作的指示信息，得到用户与发送人的协同关系特征系数的介绍，不予赘述。
情况12：用户为多个接收人中的一个，协同关系信息包括用户与其他用户的协同关系权重。
若用户为多个接收人中的一个，协同关系特征系数包括用户与发送人的协同关系特征系数，以及用户与除用户之外的接收人的协同关系特征系数。根据用户的协同关系信息得到协同关系特征系数，包括：从用户与其他用户的协同关系权重中获取用户与发送人的协同关系权重，以及用户与除用户之外的接收人的协同关系权重，根据用户与发送人的协同关系权重，得到用户与发送人的协同关系特征系数，根据用户与除用户之外的接收人的协同关系权重，得到用户与除用户之外的接收人的协同关系特征系数。
其中，根据用户与发送人的协同关系权重，得到用户与发送人的协同关系特征系数的具体描述可以参考上述情况10中的介绍，不予赘述。
其中，根据用户与除用户之外的接收人的协同关系权重，得到用户与除用户之外的接收人的协同关系特征系数的具体描述可以参考上述情况10中，对根据用户与发送人的协同关系权重，得到用户与发送人的协同关系特征系数的介绍，不予赘述。
沟通关系特征系数是根据用户的沟通关系信息得到的可以有以下4种情况：
情况13：用户为唯一的接收人，用户的沟通关系信息用于指示用户与其他用户的沟通频率。
若用户为唯一的接收人，沟通关系特征系数包括用户与发送人的沟通关系特征系数。沟通关系特征系数是根据用户的沟通关系信息得到，包括：从用户的沟通关系信息中获取用户与发送人的沟通频率，根据用户与发送人的沟通频率，得到用户与发送人的沟通关系特征系数。
一种可能的实现方式，根据用户与发送人的沟通频率，得到用户与发送人的沟通关系特征系数，包括：根据用户与发送人的沟通频率，按照沟通关系特征系数的设置规则确定用户与发送人的沟通关系特征系数。
其中，沟通关系特征系数的设置规则可以是：沟通关系特征系数和用户与发送人的沟通频率成正相关，即用户与发送人的沟通频率越大，则沟通关系特征系数越大。
示例性的，可以将用户与发送人的沟通频率作为沟通关系特征系数。以用户1的用户的沟通关系信息包括{用户1，用户2，50}；{用户1，用户3，30}；{用户1，用户4， 10}，50表示用户1与用户2的沟通频率为一天50次，30表示用户1与用户3的沟通频率为一天30次，10表示用户1与用户4的沟通频率为一天10次为例，若发送人的标识为用户2，则用户与发送人的沟通关系特征系数可以是50；若发送人的标识为用户3，则用户与发送人的沟通关系特征系数可以是30；若发送人的标识为用户4，则用户与发送人的沟通关系特征系数可以是10。
示例性的，用户与发送人的沟通频率与沟通关系特征系数存在对应关系，可以根据该对应关系确定沟通关系特征系数。例如，用户与发送人的沟通频率与沟通关系特征系数的对应关系如表5所示。表5中，用户与发送人的沟通频率为一天0-10次，则沟通关系特征系数为1，用户与发送人的沟通频率为一天11-30次，则沟通关系特征系数为2，用户与发送人的沟通频率为一天31-50次，则沟通关系特征系数为3，用户与发送人的沟通频率为一天50次以上，则沟通关系特征系数为4。以用户1的用户的沟通关系信息包括{用户1，用户2，50}；{用户1，用户3，30}；{用户1，用户4，10}，50表示用户1与用户2的沟通频率为一天50次，30表示用户1与用户3的沟通频率为一天30次，10表示用户1与用户4的沟通频率为一天10次为例，若发送人的标识为用户2，则用户与发送人的沟通关系特征系数可以是3；若发送人的标识为用户3，则用户与发送人的沟通关系特征系数可以是30；若发送人的标识为用户2，则用户与发送人的沟通关系特征系数可以是1。
表5
用户与发送人的沟通频率(次/天) 沟通关系特征系数
0-10 1
11-30 2
31-50 3
51以上 4
需要说明的是，表5仅是用户与发送人的沟通频率与沟通关系特征系数的对应关系的示例，在实际应用中，用户与发送人的沟通频率与沟通关系特征系数的对应关系还可以是其他形式，而且用户与发送人的沟通频率与沟通关系特征系数的对应关系可以是表5中的某一行、某些行、表5中的全部、或者比表5示出的更多的对应关系，本申请不进行具体限定。
需要说明的是，除了上述确定用户与发送人的沟通关系特征系数的方法之外，还可以将用户与发送人的沟通频率进行处理(例如，加一个数或者减一个数等)，并将处理后的数据作为用户与发送人的沟通关系特征系数。其中，处理后的数据应满足上述沟通关系特征系数的设置规则。
情况14：用户为唯一的接收人，用户的沟通关系信息包括沟通频率的权重。
若用户为唯一的接收人，沟通关系特征系数包括用户与发送人的沟通关系特征系数。沟通关系特征系数是根据用户的沟通关系信息得到，包括：从用户的沟通关系信息中获取用户与发送人的沟通频率的权重，根据用户与发送人的沟通频率的权重，得到用户与发送人的沟通关系特征系数。
一种可能的实现方式，根据用户与发送人的沟通频率的权重，得到用户与发送人的沟通关系特征系数，包括：将用户与发送人的沟通关系权重作为用户与发送人的沟通关 系特征系数。
示例性的，用户1的沟通关系信息可以包括{用户1，用户2，R}；{用户1，用户3，S}，其中，R为用户1与用户2的沟通关系权重，S为用户1与用户3的沟通关系权重为例，若发送人的标识为用户2，则用户与发送人的沟通关系特征系数可以是R；若发送人的标识为用户3，则用户与发送人的沟通关系特征系数可以是S。
需要说明的是，除了将用户与发送人的沟通关系权重作为用户与发送人的沟通关系特征系数之外，还可以将用户与发送人沟通关系权重进行处理(例如，加一个数或者减一个数等)，并将处理后的数据作为用户与发送人的沟通关系特征系数。其中，处理后的数据应满足上述沟通关系特征系数的设置规则。
情况15：用户为多个接收人中的一个，用户的沟通关系信息用于指示用户与其他用户的沟通频率。
若用户为多个接收人中的一个，沟通关系特征系数包括用户与发送人的沟通关系特征系数，以及用户与除用户之外的接收人的沟通关系特征系数。沟通关系特征系数是根据用户的沟通关系信息得到，包括：从用户的沟通关系信息中获取用户与发送人的沟通频率，以及用户与除用户之外的接收人的沟通频率，根据用户与发送人的沟通频率，得到用户与发送人的沟通关系特征系数，根据用户与除用户之外的接收人的沟通频率，得到用户与除用户之外的接收人的沟通关系特征系数。
其中，根据用户与发送人的沟通频率，得到用户与发送人的沟通关系特征系数的具体描述可以参考上述情况13中的介绍，不予赘述。
其中，根据用户与除用户之外的接收人的沟通频率，得到用户与除用户之外的接收人的沟通关系特征系数的具体描述可以参考上述情况13中，对根据用户与发送人的沟通频率，得到用户与发送人的沟通关系特征系数的介绍，不予赘述。
情况16：用户为多个接收人中的一个，用户的沟通关系信息包括沟通频率的权重。
若用户为多个接收人中的一个，沟通关系特征系数包括用户与发送人的沟通关系特征系数，以及用户与除用户之外的接收人的沟通关系特征系数。沟通关系特征系数是根据用户的沟通关系信息得到，包括：从用户的沟通关系信息中获取用户与发送人的沟通频率的权重，以及用户与除用户之外的接收人的沟通频率的权重，根据用户与发送人的沟通频率的权重，得到用户与发送人的沟通关系特征系数，根据用户与除用户之外的接收人的沟通频率的权重，得到用户与除用户之外的接收人的沟通关系特征系数。
其中，根据用户与发送人的沟通频率的权重，得到用户与发送人的沟通关系特征系数的具体描述可以参考上述情况14中的介绍，不予赘述。
其中，根据用户与除用户之外的接收人的沟通频率的权重，得到用户与除用户之外的接收人的沟通关系特征系数的具体描述可以参考上述情况14中，对根据用户与发送人的沟通频率的权重，得到用户与发送人的沟通关系特征系数的介绍，不予赘述。
关系特征系数除了根据接收人的标识和/或发送人的标识，以及用户的关系信息得到，还可以根据接收人的标识和/或发送人的标识、用户的关系信息以及第二用户行为反馈数据得到。
其中，第二用户行为反馈数据可以用于指示历史上与该文本的发送人相同的文本的重要性，和/或，历史上与该文本的接收人相同的文本的重要性。后续，可以根据第二用 户行为反馈数据得到第二用户行为反馈数据的影响系数，在根据上述情况1-情况16所述的方法得到关系特征系数后，可以通过第二用户行为反馈数据的影响系数对关系特征系数中的一个或多个系数进行修正，使得修正后的关系特征系数更准确。
可选的，根据第二用户行为反馈数据得到第二用户行为反馈数据的影响系数，将上述情况1-情况16所述的方法得到关系特征系数中的一个或多个系数与第二用户行为反馈数据的影响系数做加减运算，得到更准确的关系特征系数。
其中，第二用户行为反馈数据的影响系数可以用于指示历史上与该文本的发送人相同的文本的重要性，和/或，历史上与该文本的接收人相同的文本的重要性。
可选的，第二用户行为反馈数据的影响系数与文本的重要性成正相关，即在其他特征系数不变的情况下，第二用户行为反馈数据的影响系数越大，文本的重要性越高。
可选的，根据第二用户行为反馈数据得到第二用户行为反馈数据的影响系数，包括：获取第二对应关系，第二对应关系为第二用户行为反馈数据的影响系数与第二用户行为反馈数据之间的关联关系，根据第二用户行为反馈数据以及第二对应关系，确定第二用户行为反馈数据的影响系数。
其中，第二对应关系可以预设置，并存储在图2中的存储器203或服务器中。
可选的，获取第二对应关系，包括：从图2中的存储器203或服务器中获取该第二对应关系。
如表6所示，为第二用户行为反馈数据的影响系数与第二用户行为反馈数据的对应关系。表6中，若第二用户行为反馈数据指示历史上与该文本的发送人相同的文本为重要文本，则第二用户行为反馈数据的影响系数为b，若第二用户行为反馈数据指示历史上与该文本的发送人相同的文本为非重要文本，则第二用户行为反馈数据的影响系数为-b，若第二用户行为反馈数据指示历史上与该文本的接收人相同的文本为重要文本，则第二用户行为反馈数据的影响系数为c，若第二用户行为反馈数据指示历史上与该文本的接收人相同的文本为非重要文本，则第二用户行为反馈数据的影响系数为-c。
表6
Figure PCTCN2020095510-appb-000002
示例性的，以第二用户行为反馈数据的影响系数与第二用户行为反馈数据的对应关系如表6所示，关系特征系数中的上下级关系特征系数与第二用户行为反馈数据的影响系数做加减运算为例，若第二用户行为反馈数据指示历史上与该文本的发送人相同的文本为重要文本，则第二用户行为反馈数据的影响系数为b，考虑了第二用户行为反馈数据的影响系数后，新的用户与发送人的上下级关系特征系数可以为用户与发送人的上下级关系特征系数+b；若第二用户行为反馈数据指示历史上与该文本的接收人相同的文本 为非重要文本，则第二用户行为反馈数据的影响系数为-c，考虑了第二用户行为反馈数据的影响系数后，新的用户与接收人的上下级关系特征系数可以为用户与接收人的上下级关系特征系数-c。
步骤403：根据关系特征系数确定文本的重要性。
一种可能的实现方式，根据关系特征系数确定文本的重要性，包括：以关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性。具体的，可以参考上述步骤303中，以内容特征系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性的的描述，此处不再赘述。
可选的，当重要性相同的文本有多个时(例如，一级重要文本有多个时)，根据关系特征系数确定文本的重要性之后还包括：将关系特征系数中的系数做乘法运算得到文本的重要性系数；根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，将关系特征系数中的系数做乘法运算得到文本的重要性系数，包括：根据公式textImportantCoff＝upDownOranCoff*horizontalCoff*joinCoff*communicateCoff计算文本的重要性系数。其中，textImportantCoff为文本的重要性系数，upDownOranCoff为上下级关系特征系数，horizontalCoff为部门关系特征系数，joinCoff为协同关系特征系数，communicateCoff为沟通关系特征系数。
示例性的，以用户为多个接收人中的一个，用户与发送人的上下级关系特征系数为upDownOranCoff 1，用户与接收人的上下级关系特征系数为upDownOranCoff 2，用户与发送人的部门关系特征系数为horizontalCoff 1，用户与接收人的部门关系特征系数为horizontalCoff 2，用户与发送人的协同关系特征系数为joinCoff 1，用户与接收人的协同关系特征系数为joinCoff 2，用户与发送人的沟通关系特征系数为communicateCoff 1，用户与接收人的沟通关系特征系数为communicateCoff 2为例，则文本的重要性系数textImportantCoff＝upDownOranCoff 1*horizontalCoff 1*joinCoff 1*communicateCoff 1*upDownOranCoff 2*horizontalCoff 2*joinCoff 2*communicateCoff 2。
另一种可能的实现方式，根据关系特征系数确定文本的重要性，包括：将关系特征系数中的系数做乘法运算得到文本的重要性系数；根据文本的重要性系数确定文本的重要性。
将关系特征系数中的系数做乘法运算得到文本的重要性系数可以包括：根据公式textImportantCoff＝upDownOranCoff*horizontalCoff*joinCoff*communicateCoff计算文本的重要性系数。
可选的，根据文本的重要性系数确定文本的重要性，包括：若文本的重要性系数大于或等于第二阈值，则确定该文本为重要文本；或者，根据文本的重要性系数与文本的重要性的对应关系，确定文本的重要性。具体的，可以参考上述步骤303中根据文本的重要性系数确定文本的重要性的描述，此处不再赘述。
其中，第二阈值与第一阈值可以相同也可以不同。
进一步可选的，当重要性相同的文本有多个时，根据特征系数集合确定文本的重要性之后还包括：根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，在根据特征系数集合确定文本的重要性之后，将文本按照文本的重要性分类显示。
当用户更换设备时，将旧设备中的文本的重要性传输到新设备中的方法可以参考步骤303中对应的描述，此处不再赘述。
基于图4所示的方法，可以获取接收人的标识和/或发送人的标识，以及用户的关系信息，根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数，并根据关系特征系数确定文本的重要性，从而可以根据用户与接收人和/或发送人的关系确定文本的重要性。
实施例3：
用户接收文本后，识别文本的装置可以获取文本、用户的画像信息以及用户的关系信息，根据文本和用户的画像信息得到内容特征系数，根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容以及用户与接收人和/或发送人的关系确定文本的重要性。如图5所示，该识别文本的方法可以包括步骤501-步骤504。
步骤501：获取文本、用户的画像信息、以及用户的关系信息。
其中，文本可以包括文本的内容，接收人的标识以及发送人的标识。
步骤501的具体介绍可以参考上述步骤301和步骤401中对应的描述，此处不予赘述。
步骤502：根据文本和用户的画像信息得到内容特征系数。
步骤502的具体介绍可以参考上述步骤302中对应的描述，此处不予赘述。
步骤503：根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数。
步骤503的具体介绍可以参考上述步骤402中对应的描述，此处不予赘述。
需要说明的是，本申请实施例不限定步骤502以及步骤503的执行顺序，例如，可以先执行步骤502再执行步骤503，也可以先执行步骤503再执行步骤502。
步骤504：根据内容特征系数以及关系特征系数确定文本的重要性。
一种可能的实现方式，根据内容特征系数以及关系特征系数确定文本的重要性，包括：以内容特征系数以及关系特征系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性。具体的，可以参考上述步骤303中，以内容特征系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性的的描述，此处不再赘述。
可选的，当重要性相同的文本有多个时(例如，一级重要文本有多个时)，根据内容特征系数以及关系特征系数确定文本的重要性之后还包括：将内容特征系数中的N个系数全部或部分做加法运算，再与关系特征系数中的系数做乘法运算得到文本的重要性系数；根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，将内容特征系数中的N个系数全部或部分做加法运算，再与关系特征系数中的系数做乘法运算得到文本的重要性系数，包括：根据公式textImportantCoff＝(contentCoff 1+contentCoff 2+…+contentCoff N)*upDownOranCoff*horizontalCoff*joinCoff*communicateCoff计算文本的重要性系数。
另一种可能的实现方式，根据内容特征系数以及关系特征系数确定文本的重要性，包括：将内容特征系数中的N个系数全部或部分做加法运算，与关系特征系数中的系数做乘法运算得到文本的重要性系数；根据文本的重要性系数确定文本的重要性。
将内容特征系数中的N个系数全部或部分做加法运算，与关系特征系数中的系数做乘法运算得到文本的重要性系数，可以包括：根据公式textImportantCoff＝(contentCoff 1+contentCoff 2+…+contentCoff N)*upDownOranCoff*horizontalCoff*joinCoff*communicateCoff计算文本的重要性系数。
可选的，根据文本的重要性系数确定文本的重要性，包括：若文本的重要性系数大于或等于第三阈值，则确定该文本为重要文本；或者，根据文本的重要性系数与文本的重要性的对应关系，确定文本的重要性。具体的，可以参考上述步骤303中根据文本的重要性系数确定文本的重要性的描述，此处不再赘述。
其中，第三阈值与第一阈值以及第二阈值可以相同也可以不同。
进一步可选的，当重要性相同的文本有多个时，根据特征系数集合确定文本的重要性之后还包括：根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，在根据特征系数集合确定文本的重要性之后，将文本按照文本的重要性分类显示。
当用户更换设备时，将旧设备中的文本的重要性传输到新设备中的方法可以参考步骤303中对应的描述，此处不再赘述。
基于图5所示的方法，可以获取文本、用户的画像信息、接收人的标识和/或发送人的标识，以及用户的关系信息，根据文本和用户的画像信息得到内容特征系数，根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数，并根据内容特征系数以及关系特征系数确定文本的重要性，从而可以根据文本的内容以及用户与接收人和/或发送人的关系确定文本的重要性。
上述实施例1-实施例3中介绍了根据内容特征系数和/或关系特征系数确定文本的重要性的方法，除了上述方法，还可以根据消息影响范围特征系数，和/或，用户与接收人/发送人的关系系数确定文本的重要性。其中，消息影响范围特征系数可以用于指示接收人的个数，用户与接收人/发送人的关系系数可以用于指示用户与接收人和/或发送人的关系的紧密程度。
下面以根据内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数确定文本的重要性为例进行介绍。
实施例4：
用户接收文本后，识别文本的装置可以获取文本、用户的画像信息以及用户的关系信息，根据文本和用户的画像信息得到内容特征系数，根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数，根据接收人的标识得到消息影响范围特征系数，根据用户的关系信息得到用户与接收人/发送人的关系系数，并根据内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数确定文本的重要性。如图6所示，该识别文本的方法可以包括步骤601-步骤606。
步骤601：获取文本、用户的画像信息、以及用户的关系信息。
其中，文本可以包括文本的内容、接收人的标识以及发送人的标识。
步骤601的具体介绍可以参考上述步骤301和步骤401中对应的描述，此处不予赘述。
步骤602：根据文本和用户的画像信息得到内容特征系数。
步骤602的具体介绍可以参考上述步骤302中对应的描述，此处不予赘述。
步骤603：根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数。
步骤603的具体介绍可以参考上述步骤402中对应的描述，此处不予赘述。
步骤604：根据接收人的标识得到消息影响范围特征系数。
可选的，根据接收人的标识得到消息影响范围特征系数，包括：根据接收人的标识确定接收人的个数，根据接收人的个数得到消息影响范围特征系数。例如，若接收人的个数为2，则消息影响范围特征系数为2，又例如，若接收人的个数少于10，则消息影响范围特征系数为0.3，若接收人的个数大于等于10且少于30，则消息影响范围特征系数为0.8，若接收人的个数大于等于30，则消息影响范围特征系数为1。
可选的，消息影响范围特征系数与文本的重要性成非正相关，即在其他特征系数不变的情况下，消息影响范围特征系数越大，文本的重要性越低。
步骤605：根据用户的关系信息得到用户与接收人/发送人的关系系数。
若用户为发送人，用户与接收人/发送人的关系系数可以为：与用户有组织关系和/或沟通关系的接收人的关系特征系数之和，与所有接收人的关系特征系数之和的比值；或者，用户与接收人/发送人的关系系数可以为：与用户有组织关系和/或沟通关系的接收人与用户的关系权重之和，与所有接收人与用户的关系权重之和的比值。其中，关系权重可以包括上下级关系的权重、最小单元组织的权重、协同关系权重以及沟通频率的权重。
示例性的，以用户1向用户2、用户3以及用户4发送文本，用户2和用户3与用户1有组织关系以及沟通关系，用户2的关系特征系数为20，用户3的关系特征系数为33，用户4与用户1没有组织关系以及沟通关系，用户4的关系特征系数为2为例，则用户1与接收人/发送人的关系系数可以为(20+33)/(20+33+2)＝0.96。
示例性的，以用户1向用户2、用户3以及用户4发送文本，用户2和用户3与用户1有组织关系，用户2的关系权重为25，用户3的关系权重为22，用户4与用户1没有组织关系，用户4的关系权重为1为例，则用户1与接收人/发送人的关系系数可以为(25+22)/(25+22+1)＝0.98。
若用户为多个接收人中的一个，用户与接收人/发送人的关系系数可以为：与用户有组织关系和/或沟通关系的接收人以及发送人的关系特征系数之和，与所有接收人以及发送人的关系特征系数之和的比值；或者，用户与接收人/发送人的关系系数可以为：与用户有组织关系和/或沟通关系的接收人以及发送人与用户的关系权重之和，与所有接收人以及发送人与用户的关系权重之和的比值。
示例性的，以用户2向用户1、用户3以及用户4发送文本，用户2和用户3与用户1有沟通关系，用户2的关系特征系数为15，用户3的关系特征系数为26，用户4与用户1没有沟通关系，用户4的关系特征系数为2为例，则用户1与接收人/发送人的关系系数可以为(15+26)/(15+26+2)＝0.95。
示例性的，以用户2向用户1、用户3以及用户4发送文本，用户2与用户1有组织关系，用户2的关系权重为27，用户3和用户4与用户1没有组织关系，用户3的关系权重为1，用户4的关系权重为1为例，则用户1与接收人/发送人的关系系数可以为 27/(27+1+1)＝0.93。
可选的，用户与接收人/发送人的关系系数与文本的重要性成正相关，即在其他特征系数不变的情况下，用户与接收人/发送人的关系系数越大，文本的重要性越高。
需要说明的是，本申请实施例不限定步骤602-步骤605的执行顺序，例如，可以先执行步骤605，接着执行步骤604，再执行步骤603，最后执行步骤602，也可以先执行步骤604，接着执行步骤602，再执行步骤605，最后执行步骤603，还可以先执行步骤603，接着执行步骤605，再执行步骤602，最后执行步骤604。
步骤606：根据内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数确定文本的重要性。
一种可能的实现方式，根据内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数确定文本的重要性，包括：以内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性。具体的，可以参考上述步骤303中，以内容特征系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性的的描述，此处不再赘述。
可选的，当重要性相同的文本有多个时(例如，一级重要文本有多个时)，根据内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数确定文本的重要性之后还包括：将内容特征系数中的N个系数全部或部分做加法运算，与关系特征系数中的系数、用户与接收人/发送人的关系系数做乘法运算，与消息影响范围特征系数做除法运算得到文本的重要性系数；根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，将内容特征系数中的N个系数全部或部分做加法运算，与关系特征系数中的系数、用户与接收人/发送人的关系系数做乘法运算，与消息影响范围特征系数做除法运算得到文本的重要性系数，包括：根据公式textImportantCoff＝(contentCoff 1+contentCoff 2+…+contentCoff N)*upDownOranCoff*horizontalCoff*joinCoff*communicateCoff*receiveandsendCoff/textAffectRange计算文本的重要性系数。其中，receiveandsendCoff用户与接收人/发送人的关系系数，textAffectRange为消息影响范围特征系数。
另一种可能的实现方式，根据内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数确定文本的重要性，包括：将内容特征系数中的N个系数全部或部分做加法运算，与关系特征系数中的系数、用户与接收人/发送人的关系系数做乘法运算，与消息影响范围特征系数做除法运算得到文本的重要性系数；根据文本的重要性系数确定文本的重要性。
将内容特征系数中的N个系数全部或部分做加法运算，与关系特征系数中的系数、用户与接收人/发送人的关系系数做乘法运算，与消息影响范围特征系数做除法运算得到文本的重要性系数，可以包括：根据公式textImportantCoff＝(contentCoff 1+contentCoff 2+…+contentCoff N)*upDownOranCoff*horizontalCoff*joinCoff*communicateCoff*receiveandsendCoff/textAffectRange计算文本的重要性系数。
可选的，根据文本的重要性系数确定文本的重要性，包括：若文本的重要性系数大于或等于第四阈值，则确定该文本为重要文本；或者，根据文本的重要性系数与文本的重要性的对应关系，确定文本的重要性。具体的，可以参考上述步骤303中根据文本的重要性系数确定文本的重要性的描述，此处不再赘述。
其中，第四阈值与第一阈值、第二阈值以及第三阈值可以相同也可以不同。
进一步可选的，当重要性相同的文本有多个时，根据特征系数集合确定文本的重要性之后还包括：根据文本的重要性系数对用户接收或发送的文本进行排序。
可选的，在根据特征系数集合确定文本的重要性之后，将文本按照文本的重要性分类显示。
当用户更换设备时，将旧设备中的文本的重要性传输到新设备中的方法可以参考步骤303中对应的描述，此处不再赘述。
基于图6所示的方法，可以获取文本、用户的画像信息、接收人的标识和/或发送人的标识，以及用户的关系信息，根据文本和用户的画像信息得到内容特征系数，根据接收人的标识和/或发送人的标识，以及用户的关系信息得到关系特征系数，根据接收人的标识得到消息影响范围特征系数，根据用户的关系信息得到用户与接收人/发送人的关系系数，并根据内容特征系数、关系特征系数、消息影响范围特征系数，以及用户与接收人/发送人的关系系数确定文本的重要性。
下面以用户1向用户2以及用户3发送文本，识别文本的装置获取文本、用户2的关系信息以及用户2的画像信息，并根据文本、用户2的关系信息以及用户2的画像信息得到内容特征系数、关系特征系数以及消息影响范围特征系数为例，介绍识别文本的方法。
实施例5：
如图7所示，该识别文本的方法包括步骤701-步骤705。
步骤701：获取文本、用户2的关系信息以及用户2的画像信息。
其中，文本可以包括文本的内容、接收人的标识以及发送人的标识。
用户2的关系信息包括{用户1，用户2，上级，小组织1，0，30}以及{用户3，用户2，同级，小组织1，1，50}。其中，{用户1，用户2，上级，小组织1，0，30}中，上级表示用户1是用户2的上级用户，小组织1表示用户1与用户2的最小单元组织为小组织1，0表示用户1与用户2没有协同工作关系，30表示用户1与用户2一个月的沟通频率为30。{用户3，用户2，同级，小组织1，1，50}中，同级表示用户3是用户2的同级用户，小组织1表示用户3与用户2的最小单元组织为小组织1，1表示用户3与用户2有协同工作关系，50表示用户3与用户2一个月的沟通频率为50。
用户2的画像信息包括{tag 1，tag 2，…，tag 10}和{tagVal 1，tagVal 2，…，tagVal 10}。
步骤701的描述可以参考上述步骤301和步骤401中的介绍，此处不再赘述。
步骤702：根据文本和用户的画像信息，得到内容特征系数。
根据步骤302中示例4中介绍的方法，从文本中提取文本的关键词，并获取文本的关键词与{tag 1，tag 2，…，tag 10}的相似度{similarCoff 1，similarCoff 2，…，similarCoff 10}。将相似度与相似度对应的标签信息的权重相乘，并根据公式contentCoff i＝Function(x)计算内容特征系数{contentCoff 1，contentCoff 2，…，contentCoff 10}。其中，Function(x)＝exp(x)， x＝tagVal i*similarCoff i，i为大于或等于1并且小于或等于10的正整数。
步骤703：根据接收人的标识、发送人的标识以及用户的关系信息，得到关系特征系数。
根据上述步骤402中介绍的方法，可以得到用户1与用户2的关系特征系数为{6，5，0.01，30}，其中，6为用户1与用户2的上下级关系特征系数，5为用户1与用户2的部门关系特征系数，0.01为用户1与用户2的协同关系特征系数，30为用户1与用户2的沟通关系特征系数。
根据上述步骤402中介绍的方法，可以得到用户3与用户2的关系特征系数为{4，5，4，50}，其中，4为用户3与用户2的上下级关系特征系数，5为用户3与用户2的部门关系特征系数，4为用户3与用户2的协同关系特征系数，50为用户3与用户2的沟通关系特征系数。
步骤704：根据根据接收人的标识得到消息影响范围特征系数。
接收人的标识有两个，从而确定接收人的个数为2，因此，消息影响范围特征系数为2。
步骤704后，可以将上述内容特征系数、关系特征系数以及消息影响范围特征系数组合成向量[contentCoff 1，contentCoff 2，…，contentCoff 10，6，5，0.01，30，4，5，4，50，2]，共19维。
需要说明的是，本申请实施例不限制步骤702-步骤704的执行顺序，例如，可以先执行步骤703，再执行步骤702，最后执行步骤704，也可以先执行步骤704，再执行步骤702，最后执行步骤703，还可以先执行步骤702，再执行步骤704，最后执行步骤703，本申请实施例不进行具体限定。
步骤705：以内容特征系数、关系特征系数以及消息影响范围特征系数作为机器学习方法的输入数据，通过机器学习方法确定文本的重要性。
在执行步骤705前，还要训练机器学习方法的模型。该具体过程如下：
步骤1：获取历史上用户2接收到的大量文本。例如，获取历史上用户2接收到的不同文本100000条。
步骤2：将该大量文本按照文本的重要性分类标注。例如，重要文本标注为1，非重要文本标注为0。
步骤3：根据上述步骤701-步骤704，得到该大量文本(100000条文本)中，每个文本的内容特征系数、关系特征系数以及消息影响范围特征系数组合成的向量(19维)，并将100000条文本的向量按列合并，得到19*100000的矩阵。
步骤4：使用深度学习的方法训练模型。训练模型中有输入层、隐含层以及输出层。其中，可以设置4层隐含层，第一层隐含层的参数为：W1＝[12，19]，b1＝[12，1]，第二层隐含层的参数为：W2＝[8，12]，b2＝[8，1]，第三层隐含层的参数为：W3＝[4，8]，b3＝[4，1]，第四层隐含层的参数为：W4＝[2，4]，b4＝[2，1]，5层隐含层的激活函数设为relu，使用归一化指数函数(softmax)的方法得到最后分类概率，即前向传播计算公式为predict＝softmax(W4*relu(W3*relu(W2*relu(W1*inputX+b1)+b2)+b3)+b4)。接着使用交叉熵作为损失函数，学习率设为1e-5，最后使用Adam梯度下降优化器进行参数更新，训练次数设为20000次。经训练后，可以得到预测文本重要度类别的计算公式为 Y＝Index(max(predict))，其中，Y是predict数组最大元素对应的下标，当Y为0时，表示该文本为重要文本，当Y为1时，表示文本为非重要文本。
得到训练模型后，可以将步骤704后得到的用户2的内容特征系数、关系特征系数以及消息影响范围特征系数组合成的向量[contentCoff 1，contentCoff 2，…，contentCoff 10，6，5，0.01，30，4，5，4，50，2]输入训练模型，得到用户2接收到的文本的重要性。
基于图7所示的方法，可以获取文本、用户的关系信息以及用户的画像信息，根据文本以及用户的画像信息，得到内容特征系数，根据接收人的标识、发送人的标识以及用户的关系信息，得到关系特征系数，根据接收人的标识得到消息影响范围特征系数，并根据内容特征系数、关系特征系数以及消息影响范围特征系数确定文本的重要性，从而可以高效地识别文本的重要性。
上述识别文本的装置为了实现上述功能，其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所公开的实施例描述的各示例的单元及算法操作，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对识别文本的装置进行功能模块的划分，例如，可以对应各个功能划分各个功能模块，也可以将两个或两个以上的功能集成在一个处理模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
比如，以采用集成的方式划分各个功能模块的情况下，图8示出了一种识别文本的装置80的结构示意图。该识别文本的装置80包括：获取模块801、处理模块802以及确定模块803。
获取模块801，用于获取文本以及用户的画像信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示该用户相关的文本中的N个关键词，N为大于或等于1的整数。
处理模块802，用于根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示该文本的内容的重要性。
确定模块803，用于根据该内容特征系数确定该文本的重要性。
可选的，该用户的画像信息包括N个标签信息，该N个标签信息对应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。
可选的，处理模块802，具体用于获取第一用户行为反馈数据，其中，第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；处理模块802，还具体用于获取第一对应关系，其中，该第一对应关系用于指示该第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；处理模块802，还具体用于根据该第一用户行为反馈数据和该第一对应关系得到该第一用户行为反馈数据的影响系数；处理模块802，还具体用于根据该文本、该用户的画像信息以及该第一用户行为反 馈数据的影响系数得到该内容特征系数。
可选的，获取模块801，还用于获取用户的关系信息，其中，该用户的关系信息用于指示用户与组织中其他用户的层级关系；处理模块802，还用于根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；确定模块803，还用于根据该关系特征系数确定该文本的重要性。
可选的，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。
可选的，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。
可选的，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；处理模块802，具体用于用于获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；处理模块802，还具体用于获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；处理模块802，还具体用于根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；处理模块802，还具体用于根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。
可选的，该文本包括该文本的接收人的标识，处理模块802，还用于根据该文本的接收人的标识得到该消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数；确定模块803，还用于根据该消息影响范围特征系数确定该文本的重要性。
可选的，处理模块802，还用于根据该用户的关系信息得到用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度；确定模块803，还用于根据该用户与接收人/发送人的关系系数确定该文本的重要性。
可选的，确定模块803，具体用于以该内容特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，确定模块803，具体用于以该内容特征系数和该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，确定模块803，具体用于以该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，确定模块803，具体用于以该内容特征系数、该关系特征系数、该消息影响范围特征系数以及该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，当重要性相同的文本有多个时，该内容特征系数包括N个系数，如图9所示，识别文本的装置80还包括：排序模块804；处理模块802，还用于将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数；排序模块804，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，当重要性相同的文本有多个时，该内容特征系数包括N个系数，如图9所示，识别文本的装置80还包括：排序模块804；处理模块802，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；排序模块804，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，当重要性相同的文本有多个时，该内容特征系数包括N个系数，如图9所示，识别文本的装置80还包括：排序模块804；处理模块802，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；排序模块804，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，当重要性相同的文本有多个时，该内容特征系数包括N个系数，如图9所示，识别文本的装置80还包括：排序模块804；处理模块802，还用于将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；排序模块804，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，确定模块803，具体用于将该内容特征系数中的N个系数全部或部分做加法运算得到该文本的重要性系数系数；确定模块803，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，确定模块803，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；确定模块803，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，确定模块803，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；确定模块803，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，确定模块803，具体用于将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；确定模块803，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，当重要性相同的文本有多个时，排序模块804，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，如图10所示，文本识别的装置80还包括：显示模块805；显示模块805，用于将该文本按照该文本的重要性分类显示。
其中，上述方法实施例涉及的各操作的所有相关内容均可以援引到对应功能模块的功能描述，在此不再赘述。
在本实施例中，该识别文本的装置80以采用集成的方式划分各个功能模块的形式来呈现。这里的“模块”可以指特定ASIC，电路，执行一个或多个软件或固件程序的处理器和存储器，集成逻辑电路，和/或其他可以提供上述功能的器件。在一个简单的实施例中，本领域的技术人员可以想到该识别文本的装置80可以采用图2所示的形式。
比如，图2中的存储器203中存储的计算机执行指令，处理器201可以通过调用存储器203中存储的计算机执行指令，执行下述过程：从存储器203或服务器获取文本以及用户的画像信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示与该文本的重要性相关的信息；根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示文本的内容的重要性；根据内容特征系数确定该文本的重要性。
示例性的，图10中的获取模块801、处理模块802、确定模块803、排序模块804和显示模块804的功能/实现过程可以通过图2中的处理器201调用存储器203中存储的计算机执行指令来实现。或者，图10中的获取模块801、处理模块802、确定模块803和排序模块804的功能/实现过程可以通过图2中的处理器201调用存储器203中存储的计算机执行指令来实现，图10中的显示模块805的功能/实现过程可以通过图2中的输出设备205来实现。
由于本实施例提供的识别文本的装置80可执行上述的识别文本的方法，因此其所能获得的技术效果可参考上述方法实施例，在此不再赘述。
比如，以采用集成的方式划分各个功能模块的情况下，图11示出了一种识别文本的装置110的结构示意图。该识别文本的装置110包括：获取模块1101、处理模块1102以及确定模块1103。
获取模块1101，用于获取文本以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的关系信息用于指示该用户与组织中其他用户的层级关系。
处理模块1102，用于根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系。
确定模块1103，用于根据该关系特征系数确定该文本的重要性。
可选的，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。
可选的，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上 下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。
可选的，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；处理模块1102，具体用于获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；处理模块1102，还具体用于获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据和该第二用户行为反馈数据的影响系数之间的关联关系；处理模块1102，还具体用于根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；处理模块1102，还具体用于根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。
可选的，确定模块1103，具体用于以该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，当重要性相同的文本有多个时，如图12所示，识别文本的装置110还包括：排序模块1104；处理模块1102，还用于将该关系特征系数中的系数做乘法运算得到该文本的重要性系数；排序模块1104，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，确定模块1103，具体用于将该关系特征系数中的系数做乘法运算得到该文本的重要性系数；确定模块1103，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，当重要性相同的文本有多个时，排序模块1104，还用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，如图13所示，该识别文本的装置110还包括：显示模块1105；显示模块1105，用于将该文本按照该文本的重要性分类显示。
其中，上述方法实施例涉及的各操作的所有相关内容均可以援引到对应功能模块的功能描述，在此不再赘述。
在本实施例中，该识别文本的装置110以采用集成的方式划分各个功能模块的形式来呈现。这里的“模块”可以指特定ASIC，电路，执行一个或多个软件或固件程序的处理器和存储器，集成逻辑电路，和/或其他可以提供上述功能的器件。在一个简单的实施例中，本领域的技术人员可以想到该识别文本的装置110可以采用图2所示的形式。
比如，图2中的存储器203中存储的计算机执行指令，处理器201可以通过调用存储器203中存储的计算机执行指令，执行下述过程：从存储器203或服务器获取文本以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的关系信息用于指示用户与组织中其他用户的层级关系；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系；根据该关系特征系数确定该文本的重要性。
示例性的，图13中的获取模块1101、处理模块1102、确定模块1103、排序模块1104和显示模块1104的功能/实现过程可以通过图2中的处理器201调用存储器203中 存储的计算机执行指令来实现。或者，图13中的获取模块1101、处理模块1102、确定模块1103和排序模块1104的功能/实现过程可以通过图2中的处理器201调用存储器203中存储的计算机执行指令来实现，图13中的显示模块1105的功能/实现过程可以通过图2中的输出设备205来实现。
由于本实施例提供的识别文本的装置110可执行上述的识别文本的方法，因此其所能获得的技术效果可参考上述方法实施例，在此不再赘述。
比如，以采用集成的方式划分各个功能模块的情况下，图14示出了一种识别文本的装置140的结构示意图。该识别文本的装置140包括：获取模块1401、处理模块1402以及确定模块1403。
获取模块1401，用于获取文本、用户的画像信息、以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示该用户相关的文本中的N个关键词，N为大于或等于1的整数，该用户的关系信息用于指示该用户与组织中其他用户的层级关系。
处理模块1402，用于根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示文本的内容的重要性。
处理模块1402，还用于根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系。
确定模块1403，用于根据该内容特征系数以及该关系特征系数确定该文本的重要性。
可选的，该用户的画像信息包括N个标签信息，该N个标签信息对应指示该N个关键词；或者，该用户的画像信息包括该N个标签信息，以及该N个标签信息中每个标签信息的权重。
可选的，处理模块1402，具体用于获取第一用户行为反馈数据，其中，第一用户行为反馈数据用于指示历史上与该文本内容相似度高的文本的重要性；处理模块1402，还具体用于获取第一对应关系，其中，该第一对应关系用于指示该第一用户行为反馈数据和该第一用户行为反馈数据的影响系数之间的关联关系；处理模块1402，还具体用于根据该第一用户行为反馈数据、以及该第一对应关系得到该第一用户行为反馈数据的影响系数；处理模块1402，还具体用于根据该文本、该用户的画像信息以及该第一用户行为反馈数据的影响系数得到该内容特征系数。
可选的，该用户的关系信息包括该用户的组织关系信息以及该用户的沟通关系信息，该用户的组织关系信息包括该用户的上下级关系信息、该用户的部门关系信息以及协同关系信息；该用户的上下级关系信息用于指示该用户与其他用户的上下级关系；该用户的部门关系信息用于指示该用户与其他用户的最小单元组织，该最小单元组织为该用户与其他用户所在的相同组织中，用户数量最少的组织；该协同关系信息用于指示该用户与其他用户是否协同工作；该用户的沟通关系信息用于指示该用户与其他用户的沟通频率。
可选的，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；该上下级关系特征系数是根据该用户的上下级关系信息得到的；该部门关系特征系数是根据 该用户的部门关系信息得到的；该协同关系特征系数是根据该协同关系信息得到的；该沟通关系特征系数是根据该用户的沟通关系信息得到的。
可选的，该关系特征系数包括组织关系特征系数以及沟通关系特征系数，该组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数；处理模块1402，具体用于获取第二用户行为反馈数据，其中，第二用户行为反馈数据用于指示历史上与该文本的发送人相同的文本的重要性；处理模块1402，还具体用于获取第二对应关系，其中，该第二对应关系用于指示该第二用户行为反馈数据，以及该第二用户行为反馈数据的影响系数之间的关联关系；处理模块1402，还具体用于根据该第二用户行为反馈数据以及该第二对应关系，得到该第二用户行为反馈数据的影响系数；处理模块1402，还具体用于根据该文本、该用户的关系信息以及该第二用户行为反馈数据的影响系数得到该关系特征系数。
可选的，处理模块1402，还用于根据该文本得到该消息影响范围特征系数，其中，该消息影响范围特征系数用于指示该文本的接收人的个数；确定模块1403，还用于根据该消息影响范围特征系数确定该文本的重要性。
可选的，处理模块1402，还用于根据该用户的关系信息得到该用户与接收人/发送人的关系系数，其中，该用户与接收人/发送人的关系系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系的紧密程度；确定模块1403，还用于根据该用户与接收人/发送人的关系系数确定该文本的重要性。
可选的，确定模块1403，具体用于以该内容特征系数以及该关系特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，确定模块1403，具体用于以该内容特征系数、该关系特征系数和该消息影响范围特征系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，确定模块1403，具体用于以该内容特征系数、该关系特征系数、该消息影响范围特征系数和该用户与接收人/发送人的关系系数作为机器学习方法的输入数据，通过该机器学习方法确定该文本的重要性。
可选的，当重要性相同的文本有多个时，该内容特征系数包括N个系数，如图15所示，该识别文本的装置140还包括：排序模块1404；处理模块1402，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；排序模块1404，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，当重要性相同的文本有多个时，该内容特征系数包括N个系数，如图15所示，该识别文本的装置140还包括：排序模块1404；处理模块1402，还用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；排序模块1404，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，当重要性相同的文本有多个时，该内容特征系数包括N个系数，如图15所示，该识别文本的装置140还包括：排序模块1404；处理模块1402，还用于将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关 系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；排序模块1404，用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，确定模块1403，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算得到该文本的重要性系数；确定模块1403，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，确定模块1403，具体用于将该内容特征系数中的系数之和与该关系特征系数中的系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；确定模块1403，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，确定模块1403，具体用于将该内容特征系数中的系数之和，与该关系特征系数中的系数以及该用户与接收人/发送人的关系系数做乘法运算，与该消息影响范围特征系数做除法运算得到该文本的重要性系数；确定模块1403，还具体用于根据该文本的重要性系数确定该文本的重要性。
可选的，当重要性相同的文本有多个时，排序模块1404，还用于根据该文本的重要性系数对该用户接收或发送的文本进行排序。
可选的，如图16所示，该识别文本的装置140还包括：显示模块1405；该显示模块1405，用于将该文本按照该文本的重要性分类显示。
其中，上述方法实施例涉及的各操作的所有相关内容均可以援引到对应功能模块的功能描述，在此不再赘述。
在本实施例中，该识别文本的装置140以采用集成的方式划分各个功能模块的形式来呈现。这里的“模块”可以指特定ASIC，电路，执行一个或多个软件或固件程序的处理器和存储器，集成逻辑电路，和/或其他可以提供上述功能的器件。在一个简单的实施例中，本领域的技术人员可以想到该识别文本的装置140可以采用图2所示的形式。
比如，图2中的存储器203中存储的计算机执行指令，处理器201可以通过调用存储器203中存储的计算机执行指令，执行下述过程：从存储器203或服务器获取文本、用户的画像信息、以及用户的关系信息，其中，该文本为该用户接收或者发送的文本，该用户的画像信息用于指示与该文本的重要性相关的信息，该用户的关系信息用于指示用户与组织中其他用户的层级关系；根据该文本以及该用户的画像信息得到内容特征系数，其中，该内容特征系数用于指示文本的内容的重要性；根据该文本以及该用户的关系信息得到关系特征系数，其中，该关系特征系数用于指示该用户与该文本的接收人和/或该文本的发送人的关系。根据特征系数集合确定该文本的重要性，其中，该特征系数集合包括该内容特征系数以及该关系特征系数。
示例性的，图16中的获取模块1401、处理模块1402、确定模块1403、排序模块1404和显示模块1405的功能/实现过程可以通过图2中的处理器201调用存储器203中存储的计算机执行指令来实现。或者，图16中的获取模块1401、处理模块1402、确定模块1403和排序模块1404的功能/实现过程可以通过图2中的处理器201调用存储器203中存储的计算机执行指令来实现，图16中的显示模块1405的功能/实现过程可以通过图2中的输出设备205来实现。
由于本实施例提供的识别文本的装置140可执行上述的识别文本的方法，因此其所能获得的技术效果可参考上述方法实施例，在此不再赘述。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件程序实现时，可以全部或部分地以计算机程序产品的形式来实现。该计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或者数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line，DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可以用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)，光介质(例如，DVD)、或者半导体介质(例如固态硬盘(solid state disk，SSD))等。
尽管在此结合各实施例对本申请进行了描述，然而，在实施所要求保护的本申请过程中，本领域技术人员通过查看所述附图、公开内容、以及所附权利要求书，可理解并实现所述公开实施例的其他变化。在权利要求中，“包括”(comprising)一词不排除其他组成部分或步骤，“一”或“一个”不排除多个的情况。单个处理器或其他单元可以实现权利要求中列举的若干项功能。相互不同的从属权利要求中记载了某些措施，但这并不表示这些措施不能组合起来产生良好的效果。
尽管结合具体特征及其实施例对本申请进行了描述，显而易见的，在不脱离本申请的精神和范围的情况下，可对其进行各种修改和组合。相应地，本说明书和附图仅仅是所附权利要求所界定的本申请的示例性说明，且视为已覆盖本申请范围内的任意和所有修改、变化、组合或等同物。显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。

Claims (24)

  1. 一种识别文本的装置，其特征在于，所述装置包括：处理器和存储器，所述存储器存储有软件程序，所述处理器用于调用所述存储器中的软件程序执行下述过程：
    从所述存储器或服务器中获取文本以及用户的画像信息，其中，所述文本为所述用户接收或者发送的文本，所述用户的画像信息用于指示与所述用户相关的文本中的N个关键词，N为大于或等于1的整数，所述服务器与所述识别文本的装置连接，所述服务器中存储有所述文本以及所述用户的画像信息；
    根据所述文本以及所述用户的画像信息得到内容特征系数，其中，所述内容特征系数用于指示所述文本的内容的重要性；
    根据所述内容特征系数，确定所述文本的重要性。
  2. 根据权利要求1所述的装置，其特征在于，
    所述用户的画像信息包括N个标签信息，所述N个标签信息对应指示所述N个关键词；或者，
    所述用户的画像信息包括所述N个标签信息，以及所述N个标签信息中每个标签信息的权重。
  3. 根据权利要求1或2所述的装置，其特征在于，
    所述根据所述文本以及所述用户的画像信息得到内容特征系数，包括：
    从所述存储器或所述服务器中获取第一用户行为反馈数据，其中，所述第一用户行为反馈数据用于指示历史上与所述文本内容相似度高的文本的重要性；
    从所述存储器或所述服务器中获取第一对应关系，其中，所述第一对应关系用于指示所述第一用户行为反馈数据与所述第一用户行为反馈数据的影响系数之间的关联关系；
    根据所述第一用户行为反馈数据和所述第一对应关系得到所述第一用户行为反馈数据的影响系数；
    根据所述文本、所述用户的画像信息以及所述第一用户行为反馈数据的影响系数得到所述内容特征系数。
  4. 根据权利要求1-3任一项所述的装置，其特征在于，所述处理器还用于调用所述存储器中的软件程序执行下述过程：
    从所述存储器或所述服务器中获取用户的关系信息，其中，所述用户的关系信息用于指示所述用户与其他用户的层级关系；
    根据所述文本以及所述用户的关系信息得到关系特征系数，其中，所述关系特征系数用于指示所述用户与所述文本的接收人和/或所述文本的发送人的关系；
    所述确定所述文本的重要性，还包括：
    根据所述关系特征系数确定所述文本的重要性。
  5. 根据权利要求4所述的装置，其特征在于，所述关系特征系数包括组织关系特征系数以及沟通关系特征系数，所述组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数。
  6. 根据权利要求1-5任一项所述的装置，其特征在于，
    所述根据所述内容特征系数确定所述文本的重要性，包括：
    以所述内容特征系数作为机器学习方法的输入数据，通过所述机器学习方法确定所述文本的重要性。
  7. 根据权利要求6所述的装置，其特征在于，所述内容特征系数包括N个系数，当重要性相同的文本有多个时，所述处理器还用于调用所述存储器中的软件程序执行下述过程：
    将所述内容特征系数中的N个系数全部或部分做加法运算得到所述文本的重要性系数；
    根据所述文本的重要性系数对所述用户接收或发送的文本进行排序。
  8. 一种识别文本的装置，其特征在于，所述装置包括：处理器和存储器，所述存储器存储有软件程序，所述处理器用于调用所述存储器中的软件程序执行下述过程：
    从所述存储器或服务器中获取文本以及用户的关系信息，其中，所述文本为所述用户接收或者发送的文本，所述用户的关系信息用于指示所述用户与其他用户的层级关系，所述服务器与所述识别文本的装置连接，所述服务器中存储有所述文本以及所述用户的关系信息；
    根据所述文本以及所述用户的关系信息得到关系特征系数，其中，所述关系特征系数用于指示所述用户与所述文本的接收人和/或所述文本的发送人的关系；
    根据所述关系特征系数确定所述文本的重要性。
  9. 根据权利要求8所述的装置，其特征在于，所述关系特征系数包括组织关系特征系数以及沟通关系特征系数，所述组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数。
  10. 根据权利要求9所述的装置，其特征在于，所述根据所述关系特征系数确定所述文本的重要性，包括：
    以所述关系特征系数作为机器学习方法的输入数据，通过所述机器学习方法确定所述文本的重要性。
  11. 根据权利要求10所述的装置，其特征在于，当重要性相同的文本有多个时，所述处理器还用于调用所述存储器中的软件程序执行下述过程：
    将所述关系特征系数中的系数做乘法运算得到所述文本的重要性系数；
    根据所述文本的重要性系数对所述用户接收或发送的文本进行排序。
  12. 一种识别文本的方法，其特征在于，所述方法包括：
    获取文本以及用户的画像信息，其中，所述文本为所述用户接收或者发送的文本，所述用户的画像信息用于指示与所述用户相关的文本中的N个关键词，N为大于或等于1的整数；
    根据所述文本以及所述用户的画像信息得到内容特征系数，其中，所述内容特征系数用于指示所述文本的内容的重要性；
    根据所述内容特征系数，确定所述文本的重要性。
  13. 根据权利要求12所述的方法，其特征在于，
    所述用户的画像信息包括N个标签信息，所述N个标签信息对应指示所述N个关键词；或者，
    所述用户的画像信息包括所述N个标签信息，以及所述N个标签信息中每个标签 信息的权重。
  14. 根据权利要求12或13所述的方法，其特征在于，所述根据所述文本以及所述用户的画像信息得到内容特征系数，包括：
    获取第一用户行为反馈数据，其中，所述第一用户行为反馈数据用于指示历史上与所述文本内容相似度高的文本的重要性；
    获取第一对应关系，其中，所述第一对应关系用于指示所述第一用户行为反馈数据与所述第一用户行为反馈数据的影响系数之间的关联关系；
    根据所述第一用户行为反馈数据和所述第一对应关系得到所述第一用户行为反馈数据的影响系数；
    根据所述文本、所述用户的画像信息以及所述第一用户行为反馈数据的影响系数得到所述内容特征系数。
  15. 根据权利要求12-14任一项所述的方法，其特征在于，所述方法还包括：
    获取用户的关系信息，其中，所述用户的关系信息用于指示用户与其他用户的层级关系；
    根据所述文本以及所述用户的关系信息得到关系特征系数，其中，所述关系特征系数用于指示所述用户与所述文本的接收人和/或所述文本的发送人的关系；
    所述确定所述文本的重要性，还包括：
    根据所述关系特征系数确定所述文本的重要性。
  16. 根据权利要求15所述的方法，其特征在于，所述关系特征系数包括组织关系特征系数以及沟通关系特征系数，所述组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数。
  17. 根据权利要求12-16任一项所述的方法，其特征在于，所述根据所述内容特征系数确定所述文本的重要性，包括：
    以所述内容特征系数作为机器学习方法的输入数据，通过所述机器学习方法确定所述文本的重要性。
  18. 根据权利要求17所述的方法，其特征在于，所述内容特征系数包括N个系数，当重要性相同的文本有多个时，所述方法还包括：
    将所述所述内容特征系数中的N个系数全部或部分做加法运算得到所述文本的重要性系数；
    根据所述文本的重要性系数对所述用户接收或发送的文本进行排序。
  19. 一种识别文本的方法，其特征在于，所述方法包括：
    获取文本以及用户的关系信息，其中，所述文本为所述用户接收或者发送的文本，所述用户的关系信息用于指示所述用户与其他用户的层级关系；
    根据所述文本以及所述用户的关系信息得到关系特征系数，其中，所述关系特征系数用于指示所述用户与所述文本的接收人和/或所述文本的发送人的关系；
    根据所述关系特征系数确定所述文本的重要性。
  20. 根据权利要求19所述的方法，其特征在于，所述关系特征系数包括组织关系特征系数以及沟通关系特征系数，所述组织关系特征系数包括上下级关系特征系数、部门关系特征系数以及协同关系特征系数。
  21. 根据权利要求20所述的方法，其特征在于，所述根据所述关系特征系数确定所述文本的重要性，包括：
    以所述关系特征系数作为机器学习方法的输入数据，通过所述机器学习方法确定所述文本的重要性。
  22. 根据权利要求21所述的方法，其特征在于，当重要性相同的文本有多个时，所述方法还包括：
    将所述关系特征系数中的系数做乘法运算得到所述文本的重要性系数；
    根据所述文本的重要性系数对所述用户接收或发送的文本进行排序。
  23. 一种计算机存储介质，其特征在于，所述计算机可读存储介质中存储有程序指令，所述程序指令运行时，以执行权利要求12-18所述的识别文本的方法。
  24. 一种计算机存储介质，其特征在于，所述计算机可读存储介质中存储有程序指令，所述程序指令运行时，以执行权利要求19-22所述的识别文本的方法。
PCT/CN2020/095510 2019-08-02 2020-06-11 识别文本的方法及装置 WO2021022900A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP20849148.0A EP3992818A4 (en) 2019-08-02 2020-06-11 METHOD AND DEVICE FOR TEXT RECOGNITION
US17/587,498 US20220156294A1 (en) 2019-08-02 2022-01-28 Text Recognition Method and Apparatus

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN201910712935 2019-08-02
CN201910712935.5 2019-08-02
CN201910944108.9 2019-09-30
CN201910944108.9A CN110880013A (zh) 2019-08-02 2019-09-30 识别文本的方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/587,498 Continuation US20220156294A1 (en) 2019-08-02 2022-01-28 Text Recognition Method and Apparatus

Publications (1)

Publication Number Publication Date
WO2021022900A1 true WO2021022900A1 (zh) 2021-02-11

Family

ID=69727740

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/095510 WO2021022900A1 (zh) 2019-08-02 2020-06-11 识别文本的方法及装置

Country Status (4)

Country Link
US (1) US20220156294A1 (zh)
EP (1) EP3992818A4 (zh)
CN (1) CN110880013A (zh)
WO (1) WO2021022900A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110880013A (zh) * 2019-08-02 2020-03-13 华为技术有限公司 识别文本的方法及装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104615608A (zh) * 2014-04-28 2015-05-13 腾讯科技（深圳）有限公司 一种数据挖掘处理系统及方法
US20150220660A1 (en) * 2012-09-03 2015-08-06 Tencent Technology (Shenzhen) Company Limited Method and apparatus for pushing network information
CN107423362A (zh) * 2017-06-20 2017-12-01 阿里巴巴集团控股有限公司 行业确定方法、对象获取方法和装置、客户端、服务器
CN107492000A (zh) * 2016-06-13 2017-12-19 阿里巴巴集团控股有限公司 一种数据处理的方法及系统
CN109325179A (zh) * 2018-09-17 2019-02-12 青岛海信网络科技股份有限公司 一种内容推广的方法及装置
CN110880013A (zh) * 2019-08-02 2020-03-13 华为技术有限公司 识别文本的方法及装置

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001069432A2 (en) * 2000-03-16 2001-09-20 Microsoft Corporation Priorities generation and management
EP1326189A3 (en) * 2001-12-12 2005-08-17 Microsoft Corporation Controls and displays for acquiring preferences, inspecting behaviour, and guiding the learning and decision policies of an adaptive communications prioritization and routing systems
US20050204009A1 (en) * 2004-03-09 2005-09-15 Devapratim Hazarika System, method and computer program product for prioritizing messages
WO2005096584A1 (en) * 2004-03-30 2005-10-13 Imencro Software Sa A filter and a method of filtering electronic messages
JP4464975B2 (ja) * 2007-01-22 2010-05-19 インターナショナル・ビジネス・マシーンズ・コーポレーション コンピュータネットワーク上の電子文書の重要度を、当該電子文書に関係付けられた他の電子文書の当該電子文書に対する批評に基づいて、計算するためのコンピュータ装置、コンピュータプログラム及び方法
US8516058B2 (en) * 2007-11-02 2013-08-20 International Business Machines Corporation System and method for dynamic tagging in email
US20090125602A1 (en) * 2007-11-14 2009-05-14 International Business Machines Corporation Automatic priority adjustment for incoming emails
US20120245925A1 (en) * 2011-03-25 2012-09-27 Aloke Guha Methods and devices for analyzing text
TWI501097B (zh) * 2012-12-22 2015-09-21 Ind Tech Res Inst 文字串流訊息分析系統和方法
US9654437B2 (en) * 2014-03-25 2017-05-16 Palo Alto Research Center Incorporated System and method for prioritizing messages based on senders and content for drivers
CN105653562B (zh) * 2014-12-02 2019-03-15 阿里巴巴集团控股有限公司 一种文本内容与查询请求之间相关性的计算方法及装置
US10505885B2 (en) * 2015-02-27 2019-12-10 Microsoft Technology Licensing, Llc Intelligent messaging
US10333881B2 (en) * 2015-11-06 2019-06-25 Facebook, Inc. Adaptive ranking of emails in news feeds
CN106201465B (zh) * 2016-06-23 2020-08-21 扬州大学 面向开源社区的软件项目个性化推荐方法
CN107104887B (zh) * 2017-06-01 2019-02-01 珠海格力电器股份有限公司 一种即时消息提醒方法、装置及其用户终端

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150220660A1 (en) * 2012-09-03 2015-08-06 Tencent Technology (Shenzhen) Company Limited Method and apparatus for pushing network information
CN104615608A (zh) * 2014-04-28 2015-05-13 腾讯科技（深圳）有限公司 一种数据挖掘处理系统及方法
CN107492000A (zh) * 2016-06-13 2017-12-19 阿里巴巴集团控股有限公司 一种数据处理的方法及系统
CN107423362A (zh) * 2017-06-20 2017-12-01 阿里巴巴集团控股有限公司 行业确定方法、对象获取方法和装置、客户端、服务器
CN109325179A (zh) * 2018-09-17 2019-02-12 青岛海信网络科技股份有限公司 一种内容推广的方法及装置
CN110880013A (zh) * 2019-08-02 2020-03-13 华为技术有限公司 识别文本的方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3992818A4

Also Published As

Publication number Publication date
EP3992818A1 (en) 2022-05-04
CN110880013A (zh) 2020-03-13
US20220156294A1 (en) 2022-05-19
EP3992818A4 (en) 2022-09-14

Similar Documents

Publication Publication Date Title
US20190138653A1 (en) Calculating relationship strength using an activity-based distributed graph
US9946783B1 (en) Methods and systems for classifying data using a hierarchical taxonomy
US10264081B2 (en) Contextual people recommendations
US11361045B2 (en) Method, apparatus, and computer-readable storage medium for grouping social network nodes
US11704583B2 (en) Machine learning and validation of account names, addresses, and/or identifiers
CN107430625B (zh) 通过集群对文档进行分类
US20130117287A1 (en) Methods and systems for constructing personal profiles from contact data
US20160232474A1 (en) Methods and systems for recommending crowdsourcing tasks
US20180255010A1 (en) High confidence digital content treatment
US10949418B2 (en) Method and system for retrieval of data
WO2011106897A1 (en) Systems and methods for conducting more reliable assessments with connectivity statistics
EP2484054A1 (en) Systems and methods for social graph data analytics to determine connectivity within a community
US20190147404A1 (en) Email streaming records
US20210350023A1 (en) Machine Learning Systems and Methods for Predicting Personal Information Using File Metadata
US20150142717A1 (en) Providing reasons for classification predictions and suggestions
WO2018205458A1 (zh) 获取目标用户的方法、装置、电子设备及介质
US20160147886A1 (en) Querying Groups of Users Based on User Attributes for Social Analytics
US9965812B2 (en) Generating a supplemental description of an entity
WO2021022900A1 (zh) 识别文本的方法及装置
CN107851263B (zh) 用于处理推荐请求的方法和推荐引擎
JP2020135673A (ja) 投稿評価システム及び方法
US20150200903A1 (en) Automatic email address input process
US20170041277A1 (en) Method for distributing a message
EP3318987B1 (en) Method and system for retrieval of data
CN117408243A (zh) 文本内容评分方法、装置、设备和介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20849148

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020849148

Country of ref document: EP

Effective date: 20220125

NENP Non-entry into the national phase

Ref country code: DE