CN109190024B - Information recommendation method and device, computer equipment and storage medium - Google Patents

Information recommendation method and device, computer equipment and storage medium Download PDF

Info

Publication number
CN109190024B
CN109190024B CN201810948488.9A CN201810948488A CN109190024B CN 109190024 B CN109190024 B CN 109190024B CN 201810948488 A CN201810948488 A CN 201810948488A CN 109190024 B CN109190024 B CN 109190024B
Authority
CN
China
Prior art keywords
user
document
browsing
browsing data
keyword
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
CN201810948488.9A
Other languages
Chinese (zh)
Other versions
CN109190024A (en
Inventor
吴壮伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
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 Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Priority to CN201810948488.9A priority Critical patent/CN109190024B/en
Priority to PCT/CN2018/125327 priority patent/WO2020037930A1/en
Publication of CN109190024A publication Critical patent/CN109190024A/en
Application granted granted Critical
Publication of CN109190024B publication Critical patent/CN109190024B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06F16/38—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the application discloses an information recommendation method and device, computer equipment and a storage medium. The method comprises the steps of obtaining first browsing data and second browsing data of a plurality of users in a first preset time period and a second preset time period; determining a plurality of user keywords and the interest degree of each user for each user keyword according to the first browsing data and the second browsing data, and generating an interest vector of each user according to the plurality of user keywords and the corresponding interest degrees; obtaining a plurality of documents to be recommended and obtaining a document keyword of each document to be recommended and a weight value of each document keyword; generating a recommendation vector of each document to be recommended according to the document keywords and the weight values of the documents to be recommended; calculating a distance value between the interest vector of the user and the recommendation vector of each document to be recommended, and pushing the document to be recommended meeting a preset condition to the user as push information according to each distance value. The method can improve the accuracy of information recommendation.

Description

Information recommendation method and device, computer equipment and storage medium
Technical Field
The present application relates to the field of computer technologies, and in particular, to an information recommendation method and apparatus, a computer device, and a storage medium.
Background
The recommendation system is an intelligent agent system which is proposed for solving the problem of information overload and can automatically recommend resources which meet the interest preference or demand of a user from a large amount of information. With the rapid development of the internet, recommendation systems have been applied in various fields, especially in the fields of e-commerce websites and the like.
Most of the existing recommendation systems are collaborative filtering systems based on user rating matrixes, and recommend news, documents and other information which may be interested in users according to the scores of the users on the browsed news and other documents in the past. However, many users generally do not have a habit of scoring browsed news after viewing document information such as news, which causes a serious sparsity of a user scoring matrix, so that accuracy and rationality of information recommendation to the user are low, and user experience is poor.
Disclosure of Invention
The application provides an information recommendation method, an information recommendation device, computer equipment and a storage medium, so as to improve the accuracy and the rationality of information recommendation.
In a first aspect, the present application provides an information recommendation method, including: acquiring first browsing data of a plurality of users in a first preset time period and second browsing data of the users in a second preset time period, wherein the first browsing data and the second browsing data are user behavior data of the users when the users browse webpages; determining a plurality of user keywords and the interest degree of each user for each user keyword according to the first browsing data and the second browsing data, and generating an interest vector corresponding to each user according to the user keywords and the interest degrees of each user for each user keyword; acquiring a plurality of documents to be recommended, and acquiring document keywords corresponding to each document to be recommended and a weight value corresponding to each document keyword based on a preset keyword information technology; generating a recommendation vector corresponding to each document to be recommended according to the document keywords corresponding to the document to be recommended and the weight value corresponding to each document keyword; and calculating a distance value between the interest vector of the user and the recommendation vector of each document to be recommended, and pushing the document to be recommended meeting a preset condition to the user as push information according to each distance value.
In a second aspect, the present application provides an information recommendation apparatus, comprising: the browsing data acquisition unit is used for acquiring first browsing data of a plurality of users in a first preset time period and second browsing data of the users in a second preset time period, wherein the first browsing data and the second browsing data are user behavior data when the users browse webpages; an interest vector generating unit, configured to determine, according to the first browsing data and the second browsing data, a plurality of user keywords and an interest level of each user for each user keyword, and generate an interest vector corresponding to each user according to the plurality of user keywords and the interest level of each user for each user keyword; the system comprises a keyword acquisition unit, a recommendation unit and a recommendation unit, wherein the keyword acquisition unit is used for acquiring a plurality of documents to be recommended and acquiring document keywords corresponding to each document to be recommended and a weight value corresponding to each document keyword based on a preset keyword information technology; the recommendation vector generation unit is used for generating a recommendation vector corresponding to each document to be recommended according to the document keywords corresponding to the document to be recommended and the weight value corresponding to each document keyword; and the recommending unit is used for calculating a distance value between the interest vector of the user and the recommending vector of each document to be recommended, and pushing the document to be recommended meeting a preset condition to the user as pushing information according to each distance value.
In a third aspect, the present application further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the information recommendation method provided in the first aspect is implemented.
In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to execute the information recommendation method provided in the first aspect.
The application provides an information recommendation method, an information recommendation device, computer equipment and a storage medium. The information recommendation method can be used for recommending the user by combining the browsing data of the user in the first preset time period and the second preset time period, so that the accuracy and the reasonability of information recommendation are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flowchart of an information recommendation method provided in an embodiment of the present application;
fig. 2 is a specific schematic flowchart of an information recommendation method according to an embodiment of the present application;
fig. 3 is a specific schematic flowchart of an information recommendation method according to an embodiment of the present application;
fig. 4 is a specific schematic flowchart of an information recommendation method provided in an embodiment of the present application
Fig. 5 is a schematic block diagram of an information recommendation apparatus according to an embodiment of the present application;
fig. 6 is a schematic block diagram of a computer device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1, fig. 1 is a schematic flow chart of an information recommendation method according to an embodiment of the present application. The information recommendation method includes steps S101 to S105.
S101, first browsing data of a plurality of users in a first preset time period and second browsing data of the users in a second preset time period are obtained, wherein the first browsing data and the second browsing data are user behavior data when the users browse webpages.
In this embodiment, the time lengths of the first preset time period and the second preset time period are different. For example, the time length of the first predetermined period may be smaller than the time length of the second predetermined period, for example, the first predetermined period is approximately 7 days, and the second predetermined period is approximately 90 days. Therefore, the first browsing data in the first preset time period is equivalent to short-term data, and the second browsing data in the second preset time period is equivalent to long-term data, so that information is recommended to a user by combining the long-term data and the short-term data, the accuracy of information recommendation is improved, and the problem of sparsity of a user scoring matrix can be solved.
In an embodiment, the specific manner of acquiring the user behavior data when the user browses the Web page, that is, acquiring the first browsing data and the second browsing data of the user, may be acquiring through a Web server log, or performing implicit acquisition through software running on a client, and the like, where the manner of acquiring the first browsing data and the second browsing data of the user is not limited.
In an embodiment, the first browsing data may include all documents browsed by a plurality of users within a first preset time period and browsing behavior parameters of the plurality of users for each document within the first preset time period, where the browsing behavior parameters may include a click parameter of each user for each document in the first browsing data, a start time and an end time of each user when browsing each document in the first browsing data, and the like. Similarly, the second browsing data also includes all documents browsed by a plurality of users in the second preset time period and browsing behavior parameters of the plurality of users for each document in the second preset time period, where the browsing behavior parameters may include a click parameter of each user for each document in the second browsing data, a start time and an end time of each user when browsing each document in the second browsing data, and so on. Of course, the first browsing data or the second browsing data may also include other data, such as a network address of the user, a URL (Uniform Resource Locator) link of the document, and the like, which is not limited herein.
S102, determining a plurality of user keywords and the interest degree of each user for each user keyword according to the first browsing data and the second browsing data, and generating an interest vector corresponding to each user according to the user keywords and the interest degrees of each user for each user keyword.
After the first browsing data and the second browsing data are obtained, a plurality of user keywords and the interest degree of each user for each user keyword are determined according to the first browsing data and the second browsing data.
Specifically, in an embodiment, as shown in fig. 2, fig. 2 is a specific schematic flowchart of an information recommendation method provided in an embodiment of the present application. In this embodiment, the first browsing data includes a plurality of documents browsed by the user in the first preset time period and a plurality of browsing behavior parameters of the user for each document in the first preset time period; the second browsing data comprises a plurality of documents browsed by the user in the second preset time period and a plurality of browsing behavior parameters of the user to each document in the second preset time period. In step S102, determining a plurality of user keywords and a degree of interest of each user for each user keyword according to the first browsing data and the second browsing data specifically includes steps S1021 to S1025.
S1021, based on a document theme generation model, obtaining a plurality of first themes corresponding to a plurality of documents in the first browsing data and a first keyword list corresponding to each first theme, and obtaining a plurality of second themes corresponding to a plurality of documents in the second browsing data and a second keyword list corresponding to each second theme, wherein the first keyword list and the second keyword list both include a plurality of theme keywords corresponding to corresponding themes and a weight value corresponding to each theme keyword.
In this embodiment, a document topic generation model (english name: late Dirichlet Allocation, abbreviated as LDA) is used to obtain a topic corresponding to each document in the first browsing data and the second browsing data and a keyword list corresponding to each topic.
Specifically, a plurality of documents in the first browsing data are input into the document theme generation model to obtain a plurality of first themes corresponding to the plurality of documents and a first keyword list corresponding to each first theme, where the first keyword list includes a plurality of theme keywords corresponding to each first theme and a weight value corresponding to each theme keyword. Similarly, a plurality of second topics corresponding to a plurality of documents in the second browsing data and a second keyword list corresponding to each second topic can be obtained, where the second keyword list includes a plurality of topic keywords corresponding to each second topic and a weight value corresponding to each topic keyword.
It should be noted that the topic keyword is a word with a preset number of weighted values arranged from large to small in a plurality of words corresponding to the corresponding topic. For example, the topic keyword is the first 10 words with higher weight values in the plurality of words corresponding to the corresponding topic.
S1022, performing union operation on the multiple topic keywords in the first browsing data and the multiple topic keywords in the second browsing data to obtain multiple user keywords.
Since the first browsing data and the second browsing data are user behavior data when the user browses the web page in different time periods, the first browsing data and the second browsing data may embody interest preferences of the user in different time periods. As time changes, the document types, document contents, and the like preferred by the user in the first preset time period and the second preset time period may be the same, and may also be different, so that a plurality of first topics in the first browsing data and a plurality of second topics in the second browsing data may have the same topic and different topics, and the topic keywords in the corresponding plurality of first topics and the topic keywords in the plurality of second topics may also have the same keywords and different keywords. In order to recommend more accurate information to the user according to the preference of the user in different time periods, in this embodiment, a union operation needs to be performed on the plurality of topic keywords in the first browsing data and the plurality of topic keywords in the second browsing data to obtain the plurality of user keywords. For example, the plurality of topic keywords in the first browsing data include "science", "Zhang Cunzhi", and the plurality of topic keywords in the second browsing data include "science", "blood pressure", and "investment financing", so that the plurality of user keywords obtained by performing the union operation include "science", "Zhang Cunzhi", "blood pressure", and "investment financing".
And S1023, based on preset calculation rules, calculating the interest degree of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameters in the first browsing data, and calculating the interest degree of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameters in the second browsing data.
After obtaining the plurality of topic keywords in the first browsing data and the second browsing data, respectively, it is necessary to calculate the interest level of each user for each topic keyword in the first browsing data and the interest level of each topic keyword in the second browsing data.
Specifically, in an embodiment, as shown in fig. 3, fig. 3 is a specific schematic flowchart of an information recommendation method provided in an embodiment of the present application. The step S1023 includes steps S10231 to S10238.
And S10231, calculating the interest level of each user in each document in the first browsing data and the interest level of each document in the second browsing data according to the browsing behavior parameters of each user, the word count of each document and the current browsing time.
The current browsing time may be understood as the time when the application program in the user terminal needs to obtain the document of interest of the user from the server. For example, the current browsing time may be a time when the user clicks an icon of an application program in the terminal, and the application program sends a request to the corresponding server to obtain a document in which the user is interested, or may be understood as a time when the user sends a request to the corresponding server to obtain a document in which the user is interested during a process of browsing a page of the application program, and the like.
Specifically, in an embodiment, as shown in fig. 4, fig. 4 is a specific schematic flowchart of an information recommendation method provided in an embodiment of the present application. In this embodiment, the browsing behavior parameter includes a click parameter I of each user on each document in the first browsing data or the second browsing data ij Starting time T of each user when browsing each document 0ij And a termination time T ij . The step S10231 includes steps S10231a to S10231i.
S10231a, acquiring all the documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document, and acquiring all the documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document.
In this embodiment, the user clicks on the parameter I for each document ij The click parameter I is used for indicating whether the user clicks and reads the document or not, and if the user clicks and conspires the document, the click parameter I of the document ij The value of (1) is 1, if the user does not click to collude the document, the click parameter I of the document ij Is 0.
Since the plurality of documents in the first browsing data are a set of all documents browsed by the plurality of users within the first preset time period, for a certain user, it may be that only a part of the documents in the first browsing data are read, and the other documents are documents read by other users, so when calculating the browsing speed of each user, all the documents browsed by each user within the first preset time period need to be acquired, that is, all the documents browsed by each user are screened from the first browsing data.
Specifically, the click parameter I can be judged ij Is 1 to filter out all the documents that each user has browsed. Of course, all the documents browsed by each user can be filtered out by whether the start time and the end time of each document are non-null values. After all the documents browsed by each user in the first preset time period are screened out, the start time and the end time corresponding to the documents also need to be acquired. Similarly, all the documents browsed by each user within the second preset time period and the starting time and the ending time of browsing each document can be obtained.
S10231b, counting the total word number and the total time consumed by all the documents browsed by each user in the first preset time period according to all the documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document.
In this embodiment, when counting the total time consumed by all the documents browsed by each user within the first preset time period, the time of each document browsed by the user is calculated. Specifically, the difference between the ending time and the starting time of each document may be calculated to obtain the time taken by the user to browse each document. Then, the time consumed by the user to browse all the documents is summed to obtain the total time consumed by the user to browse all the documents. Similarly, when counting the total words of all the documents browsed by each user, the words of each document browsed by the user may be counted first, and then the words of all the documents browsed by the user may be summed to obtain the total words of all the documents browsed by the user.
In an embodiment, in some cases, the time consumed by the user to browse the document is sometimes abnormal, for example, after the user clicks the page of the entered document, the user exits immediately, so that the time consumed to browse the document is abnormal time, and for example, after the user clicks the page of the entered document, the user does other things and stays on the page of the document for a long time, so that the time consumed to browse the document is also abnormal time. In order to accurately calculate the first browsing speed and the second browsing speed of the user, the time taken for the user to browse each document needs to be filtered before step S10231 b.
Specifically, the method further includes, before step S10231 b: calculating the time length of each user to each document browsed in the first preset time period and calculating the time length of each user to each document browsed in the second preset time period according to the starting time and the ending time of each article browsed by the user; and respectively extracting the effective duration of each user in the first preset time period and the second preset time period and the document corresponding to the effective duration through normal distribution. That is to say, the duration of each user for each document browsed in the first preset time period and the duration of each user for each document browsed in the second preset time period are calculated, and then durations between [ -3 σ, +3 σ ] in the first preset time period and the duration between the [ -3 σ, +3 σ ] in the second preset time period are extracted through normal distribution respectively and serve as valid durations in the first preset time period and the second preset time period, so that abnormal time can be removed. And finally, obtaining the effective duration of each user in the first preset time period and the second preset time period and the document corresponding to the effective duration.
Thus, step S10231b is specifically: and counting the total word number of the documents corresponding to all the effective durations of the users in the first preset time period and the total time corresponding to all the effective durations according to the documents corresponding to all the effective durations browsed by the users in the first preset time period and the corresponding effective durations. Therefore, the first browsing speed can be calculated subsequently according to the total word number of the document corresponding to the effective duration and the total time corresponding to the effective duration.
And S10231c, counting the total word number and the total time consumed by all the documents browsed by each user in the second preset time period according to all the documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document.
The total word count and the total time consumed by all the documents browsed by each user in the second preset time period can be counted according to the statistical method of the step S10231 b.
In an embodiment, after the filtering of the time spent by the user browsing each document in the second preset time period before step S10231b, step S10231b specifically includes: and counting the total word number of the documents corresponding to all the effective durations of the users in the second preset time period and the total time corresponding to all the effective durations according to the documents corresponding to all the effective durations browsed by the users in the second preset time period and the corresponding effective durations.
And S10231d, calculating a first browsing speed of each user according to the total word number and the total time of all the documents browsed by each user in the first preset time period.
Specifically, the total word count of all the documents browsed by each user in the first preset time period is divided by the total time consumed to obtain the first browsing speed of each user.
And S10231e, calculating a second browsing speed of each user according to the total word number and the total time spent by each user in all the documents browsed in the second preset time period.
Specifically, the total word count of all the documents browsed by each user in the second preset time period is divided by the total time consumed to obtain the second browsing speed of each user.
And S10231f, calculating the attention degree of each user to each document browsed in the first preset time period according to the first browsing speed of each user, the starting time and the ending time of browsing each document in the first preset time period and the word number of each document.
Specifically, the following formula (1) may be adopted to calculate the attention degree of each user to each document browsed in the first preset time period. The calculation formula (1) is as follows:
Figure BDA0001770885160000091
wherein, C ij Representing the attention degree, speed, of the ith user to the jth document browsed in the first preset time period i Indicating a first browsing speed, T, of the ith user 0ij And T ij Respectively representing the start time and the end time of the ith user in browsing the jth document, size j The number of words representing the jth document.
And S10231g, calculating the attention degree of each user to each document browsed in the second preset time period according to the second browsing speed of each user, the starting time and the ending time of browsing each document in the second preset time period and the word number of each document.
According to the above calculation formula (1), speed is calculated i And replacing the second browsing speed of the ith user, calculating the attention degree of each user to each document browsed in the second preset time period.
S10231h, calculating the interest degree of each user in each document in the first browsing data according to the attention degree of each user to each document browsed in the first preset time period, the click parameter of each user to each document in the first browsing data, the starting time of browsing each document and the current browsing time.
Specifically, the following calculation formula (2) may be employed to calculate the interest level R of each user for each document in the first browsing data. The calculation formula (2) is as follows:
Figure BDA0001770885160000092
wherein R is ij Representing the degree of interest of the ith user in the jth document, C ij Indicating the degree of interest of the ith user in the jth document, I ij The click parameter of the ith user on the jth document is shown, lambda is a time decay constant,
Figure BDA0001770885160000093
represents a time decay factor, T represents a current browsing time, T 0ij Indicating the starting time of the ith user for browsing the jth document.
And S10231i, calculating the interest degree of each user in each document in the second browsing data according to the attention degree of each user to each document browsed in the second preset time period, the click parameter of each user to each document in the second browsing data, the starting time of browsing each document and the current browsing time.
Based on the above-described calculation formula (2), the interest level of each user in each document in the second browsing data can be calculated in the same manner.
S10232, based on the preset keyword information technology, respectively obtaining the browsing keywords of each document in the first browsing data and the weight values corresponding to the browsing keywords, and obtaining the browsing keywords of each document in the second browsing data and the weight values corresponding to the browsing keywords.
The predetermined keyword information technique may be, for example, TF-IDF (full english name: term Frequency-Inverse document Frequency). And acquiring browsing keywords of each document in the first browsing data and a weight value corresponding to each browsing keyword based on TF-IDF. Similarly, the browsing keyword of each document in the second browsing data and the weight value corresponding to each browsing keyword may be obtained based on TF-IDF.
For example, the first 10 browsing keywords with a larger weight value of each document in the first browsing data and the weight value corresponding to each browsing keyword are obtained based on the TF-IDF, and the first 10 browsing keywords with a larger weight value of each document in the second browsing data and the weight value corresponding to each browsing keyword are obtained.
S10233, calculating the probability of each document in the first browsing data on each first topic according to the browsing keywords of each document in the first browsing data, the weight value corresponding to each browsing keyword, a plurality of first topics in the first browsing data and a first keyword list corresponding to each first topic.
For example, the a document in the first browsing data corresponds to two browsing keywords and corresponding weight values, which are respectively represented as (china, 0.4) and (national flag, 0.6). Assume that the number of the first subjects is two, wherein one of the first subjects includes a subject keyword and the corresponding weight value is represented as (china, 0.6), and the other of the first subjects includes a subject keyword and the corresponding weight value is represented as (china, 0.7) and (national flag, 0.3), respectively. Then the probability of the a document on the two first topics is calculated as: the probabilities of this browsing keyword "china" on the two first topics were calculated to be 0.4 × 0.6=0.24 and 0.4 × 0.7=0.28, respectively. Then, the probabilities of the "national flag" browsing keyword on the two first topics are respectively calculated to be 0.6 × 0 =0and 0.6 × 0.3=0.18. Then the probability of the A document on the first subject is calculated to be 0.24+0=0.24 and the probability on the second first subject is calculated to be 0.28+0.18=0.46. The probability of each document on a different first topic can be calculated according to the calculation method.
S10234, calculating the probability of each document in the second browsing data on each second topic according to the browsing keywords of each document in the second browsing data, the weight values corresponding to the browsing keywords, a plurality of second topics in the second browsing data and a second keyword list corresponding to the second topics.
The probability of each document in the second browsing data on each second topic may be calculated according to the same calculation method as in step S10233.
S10235, obtaining the interest degree of each user in each first topic according to the interest degree of each user in each document in the first browsing data and the probability of each document in the first browsing data on each first topic.
In this embodiment, it is assumed that n documents are included in the first browsing data, and m first topics are included in the first browsing data. The interest degree of the ith user to the jth document is expressed as R ij Wherein j is an integer from 1 to n. The probability of the jth document on the kth first subject is represented as P jk Wherein k is an integer of 1 to m. Thus, the interest degree Q of the ith user on the kth first subject ik The expression is shown in formula (3):
Figure BDA0001770885160000111
the interest level of each user in each first topic can be calculated in turn by the above formula (3).
S10236, obtaining the interest degree of each user in each second topic according to the interest degree of each user in each document in the second browsing data and the probability of each document in the second browsing data on each second topic.
According to the above formula (3), the interest level of each user in each second topic can be calculated in the same way.
S10237, calculating the interest degree of each user for each topic keyword in the first browsing data according to the interest degree of each user for each first topic, a plurality of topic keywords in a first keyword list of each first topic and a weight value corresponding to each topic keyword.
Since each first topic has a corresponding topic keyword and a weight value corresponding to the topic keyword, the interest degree of each user in different topic keywords can be calculated. Specifically, the interest degree of each user in each first topic is multiplied by the weighted value of the topic keyword of the first topic to obtain the interest degree of each topic keyword in each first topic, then the interest degrees of the same topic keywords in a plurality of first topics are summed to obtain the interest degree of each user in each topic keyword in the first browsing data, and the interest degree of each user in each topic keyword in the first browsing data is adopted
Figure BDA0001770885160000112
And the interest degree of the ith user to the jth topic keyword in the first browsing data is represented.
S10238, calculating the interest degree of each user for each topic keyword in the second browsing data according to the interest degree of each user for each second topic, a plurality of topic keywords in a second keyword list of each second topic and the weight value corresponding to each topic keyword.
According to the same calculation method in step S10237, the interest level of each user in each topic keyword in the second browsing data can be calculated, and the method adopts
Figure BDA0001770885160000113
And the interest degree of the ith user to the jth topic keyword in the second browsing data is represented.
S1024, acquiring a first weight value corresponding to the first preset time period and a second weight value corresponding to the second preset time period.
The first weight value and the second weight value are preset and respectively used for representing the importance degree of first browsing data in a first preset time period and second browsing data in a second preset time period to the finally recommended information. This first weight value and second weight value can set up according to actual demand, for example, can set up that first weight value and second weight value are 0.5, perhaps sets up that first weight value is 0.6, and the second weight value is 0.4.
And S1025, taking the first weighted value as the weight of the interest degree of each topic keyword in the first browsing data and taking the second weighted value as the weight of the interest degree of each topic keyword in the second browsing data, and calculating the interest degree of each user on each user keyword according to a preset calculation formula. Wherein, the preset calculation formula is the following formula (4):
Figure BDA0001770885160000121
wherein, F ij Representing the interest degree of the ith user to the jth user keyword, wherein x is a first weight value, y is a second weight value,
Figure BDA0001770885160000122
indicates the interest degree of the ith user in the jth user keyword in the first browsing data, and then gets the result of the judgment>
Figure BDA0001770885160000123
And indicating the interest degree of the ith user in the keywords of the jth user in the second browsing data.
For example, the plurality of topic keywords in the first browsing data include "science" and "zhangbaizhi", and step S1023 calculates that the user a has a degree of interest of 0.2 in "science" and a degree of interest of 0.8 in "zhangbaizhi". The plurality of topic keywords in the second browsing data include "department", "blood pressure", and "investment financing", and step S1023 calculates that the user a has a degree of interest in "department" of 0.4, a degree of interest in "blood pressure" of 0.5, and a degree of interest in "investment financing" of 0.1. It is assumed that the first weight value x is 0.4 and the second weight value y is 0.6. In this way, among the plurality of user keywords, the interest level of the user a in "science ratio" is: f =0.4 + 0.2+0.6 + 0.4=0.32, and similarly, the degree of interest of the A user in Zhang Baizhi is: f =0.4 + 0.8+0.6 +0= 0.32, and so on, the interest levels of the "blood pressure" and the "investment financing" can be obtained in turn according to the preset calculation formula.
After determining a plurality of user keywords and the interest level of each user for each user keyword according to the first browsing data and the second browsing data in step S102, inputting the plurality of user keywords and the interest level of each user for each user keyword into a word vector model to generate an interest vector of a preset dimension corresponding to each user. For example, the interest vector may be a 256-dimensional vector.
S103, obtaining a plurality of documents to be recommended, and obtaining document keywords corresponding to each document to be recommended and a weight value corresponding to each document keyword based on a preset keyword information technology.
When a document needs to be recommended to a user, a plurality of documents to be recommended are obtained, wherein the plurality of documents to be recommended may be the latest updated documents. And then, extracting keywords of each acquired document to be recommended based on a preset keyword information technology, such as a TF-IDF technology, so as to obtain a plurality of document keywords corresponding to each document to be recommended, and meanwhile, obtaining a weight value corresponding to each document keyword, namely a TF-IDF value corresponding to each document keyword.
S104, generating a recommendation vector corresponding to each document to be recommended according to the document keywords corresponding to the document to be recommended and the weight value corresponding to each document keyword.
After the document keywords corresponding to each document to be recommended and the weight values corresponding to the document keywords are obtained, recommendation vectors corresponding to each document to be recommended are generated according to the document keywords and the corresponding weight values.
Specifically, in an embodiment, a plurality of document keywords corresponding to each document to be recommended and a weight value corresponding to each document keyword may be input into the word vector model to generate a recommendation vector of a preset dimension. For example, the recommendation vector may be a 256-dimensional vector.
S105, calculating a distance value between the interest vector of the user and the recommendation vector of each document to be recommended, and pushing the document to be recommended meeting a preset condition to the user as push information according to each distance value.
After the interest vector of each user is obtained through the step S102 and the recommendation vector corresponding to each document to be recommended is obtained through the step S104, a distance value between the interest vector of the user and the recommendation vector of each document to be recommended is calculated.
Specifically, in an embodiment, a distance value between the interest vector of the user and the recommendation vector of each document to be recommended may be calculated by a preset cosine similarity calculation formula. The preset cosine similarity calculation formula may be, for example, the following formula (5):
Figure BDA0001770885160000131
in the preset cosine similarity calculation formula shown in formula (5), the
Figure BDA0001770885160000133
Represents an interest vector of the user, which &>
Figure BDA0001770885160000132
And the cos theta represents a distance value between the interest vector of the user and the recommendation vector of the document to be recommended.
After the distance value between the interest vector of the user and the recommendation vector of each document to be recommended is calculated, a plurality of distance values are obtained, and then the document to be recommended meeting the preset conditions is pushed to the user as push information according to each distance value.
Specifically, in an embodiment, the document to be recommended of the recommendation vector corresponding to the minimum distance value of the preset number of distance values may be recommended to the user as recommendation information. And arranging the distance values in a descending order, and recommending the document to be recommended of the recommendation vector corresponding to the distance values of the preset number to the user as recommendation information. The preset number can be set according to actual requirements, for example, the preset number can be set to 5, and then documents to be recommended of recommendation vectors corresponding to the minimum 5 distance values are obtained from the distance values and serve as recommendation information, and the 5 recommendation information is recommended to corresponding users, so that information recommendation is completed.
The information recommendation method in the embodiment can recommend the user by combining the browsing data of the user in the first preset time period and the second preset time period, so that the accuracy and the rationality of information recommendation are improved.
The embodiment of the application also provides an information recommendation device, and the information recommendation device is used for executing any one of the information recommendation methods. Specifically, please refer to fig. 5, wherein fig. 5 is a schematic block diagram of an information recommendation device according to an embodiment of the present application. The information recommendation apparatus 300 includes a browsing data acquisition unit 301, an interest vector generation unit 302, a keyword acquisition unit 303, a recommendation vector generation unit 304, and a recommendation unit 305.
The browsing data acquiring unit 301 is configured to acquire first browsing data of a plurality of users in a first preset time period and second browsing data of the plurality of users in a second preset time period, where the first browsing data and the second browsing data are user behavior data of the plurality of users when browsing a webpage.
An interest vector generating unit 302, configured to determine a plurality of user keywords and an interest level of each user for each user keyword according to the first browsing data and the second browsing data, and generate an interest vector corresponding to each user according to the plurality of user keywords and the interest level of each user for each user keyword.
In an embodiment, the first browsing data includes a plurality of documents browsed by the user in the first preset time period and a plurality of browsing behavior parameters of the user for each document in the first preset time period; the second browsing data comprises a plurality of documents browsed by the user in the second preset time period and a plurality of browsing behavior parameters of the user to each document in the second preset time period. When determining a plurality of user keywords and a degree of interest of each user for each user keyword according to the first browsing data and the second browsing data, the interest vector generating unit 302 is specifically configured to obtain, based on a document topic generation model, a plurality of first topics corresponding to a plurality of documents in the first browsing data and a first keyword list corresponding to each first topic, and obtain a plurality of second topics corresponding to a plurality of documents in the second browsing data and a second keyword list corresponding to each second topic, where the first keyword list and the second keyword list both include a plurality of topic keywords corresponding to corresponding topics and a weight value corresponding to each topic keyword; performing union operation on the plurality of topic keywords in the first browsing data and the plurality of topic keywords in the second browsing data to obtain a plurality of user keywords; based on a preset calculation rule, calculating the interest degree of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameters in the first browsing data, and calculating the interest degree of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameters in the second browsing data; acquiring a first weight value corresponding to the first preset time period and a second weight value corresponding to the second preset time period; and taking the first weight value as the weight of the interest degree of each topic keyword in the first browsing data and the second weight value as the weight of the interest degree of each topic keyword in the second browsing data, and calculating the interest degree of each user on each user keyword according to a preset calculation formula.
Further, in an embodiment, the interest vector generating unit 302, when calculating the interest level of each user in each topic keyword in the first browsing data according to the document and browsing behavior parameters in the first browsing data and calculating the interest level of each user in each topic keyword in the second browsing data according to the document and browsing behavior parameters in the second browsing data, respectively, based on a preset calculation rule, is specifically configured to calculate the interest level of each user in each document in the first browsing data and the interest level of each document in the second browsing data according to the browsing behavior parameters of each user, the word count of each document, and the current browsing time; respectively acquiring browsing keywords of each document in the first browsing data and a weight value corresponding to each browsing keyword based on the preset keyword information technology, and acquiring browsing keywords of each document in the second browsing data and a weight value corresponding to each browsing keyword; calculating the probability of each document in the first browsing data on each first topic according to the browsing keyword of each document in the first browsing data and the weight value corresponding to each browsing keyword, and the plurality of first topics in the first browsing data and the first keyword list corresponding to each first topic; calculating the probability of each document in the second browsing data on each second topic according to the browsing keyword of each document in the second browsing data and the weight value corresponding to each browsing keyword, and a plurality of second topics in the second browsing data and a second keyword list corresponding to each second topic; obtaining the interest degree of each user in each first topic according to the interest degree of each user in each document in the first browsing data and the probability of each document in the first browsing data on each first topic; obtaining the interest degree of each user in each second topic according to the interest degree of each user in each document in the second browsing data and the probability of each document in the second browsing data on each second topic; calculating the interest degree of each user in each topic keyword in the first browsing data according to the interest degree of each user in each first topic, a plurality of topic keywords in a first keyword list of each first topic and a weight value corresponding to each topic keyword; and calculating the interest degree of each user for each topic keyword in the second browsing data according to the interest degree of each user for each second topic, a plurality of topic keywords in a second keyword list of each second topic and a weight value corresponding to each topic keyword.
Further, in an embodiment, the browsing behavior parameters include a click parameter of each of the users on each of the documents in the first browsing data or the second browsing data, a start time and an end time of each of the users when browsing each of the documents. The interest vector generating unit 302, when calculating the interest level of each user in each document in the first browsing data and the interest level of each document in the second browsing data according to the browsing behavior parameters, the word count of each document, and the current browsing time of each user, is specifically configured to obtain all documents browsed by each user in the first preset time period and a start time and an end time of browsing each document, and obtain all documents browsed by each user in the second preset time period and a start time and an end time of browsing each document; counting the total word number and the total time consumed by all the documents browsed by each user in the first preset time period according to all the documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document; counting the total word number and the total time consumed by all the documents browsed by each user in the second preset time period according to all the documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document; calculating a first browsing speed of each user according to the total word number and the total time of all the documents browsed by each user in the first preset time period; calculating a second browsing speed of each user according to the total word number and the total time consumed by all the documents browsed by each user in the second preset time period; calculating the attention degree of each user to each document browsed in the first preset time period according to the first browsing speed of each user, the starting time and the ending time of browsing each document in the first preset time period and the word number of each document; calculating the attention degree of each user to each document browsed in the second preset time period according to the second browsing speed of each user, the starting time and the ending time of browsing each document in the second preset time period and the word number of each document; calculating the interest degree of each user in each document in the first browsing data according to the attention degree of each user to each document browsed in the first preset time period, the click parameter of each user to each document in the first browsing data, the starting time of browsing each document and the current browsing time; and calculating the interest degree of each user in each document in the second browsing data according to the attention degree of each user to each document browsed in the second preset time period, the click parameter of each user to each document in the second browsing data, the starting time of browsing each document and the current browsing time.
Further, in an embodiment, the interest vector generating unit 302 is further configured to calculate a duration of each document browsed by each user in the first preset time period according to the start time and the end time of each article browsed by the user, and calculate a duration of each document browsed by each user in the second preset time period according to the start time and the end time of each article browsed by the user; and respectively extracting the effective duration of each user in the first preset time period and the second preset time period and the document corresponding to the effective duration through normal distribution.
In this way, the interest vector generating unit 302, when counting the total word count and the total time consumed by each user for all the documents browsed by the user in the first preset time period according to all the documents browsed by the user in the first preset time period and the starting time and the ending time of browsing each document, is specifically configured to count the total word count and the total time corresponding to all the effective durations of the documents browsed by each user in the first preset time period according to all the documents corresponding to all the effective durations of the user in the first preset time period and the corresponding effective durations;
similarly, when performing statistics on the total word count and the total consumed time of all documents browsed by each user in the second preset time period according to all documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document, the interest vector generating unit 302 is specifically configured to count the total word count and the total time corresponding to all documents corresponding to all effective durations browsed by each user in the second preset time period according to all documents corresponding to all effective durations and corresponding effective durations browsed by each user in the second preset time period.
The keyword obtaining unit 303 is configured to obtain a plurality of documents to be recommended, and obtain, based on a preset keyword information technology, a document keyword corresponding to each document to be recommended and a weight value corresponding to each document keyword.
A recommendation vector generating unit 304, configured to generate a recommendation vector corresponding to each to-be-recommended document according to the document keywords corresponding to the to-be-recommended document and the weight values corresponding to each of the document keywords.
The recommending unit 305 is configured to calculate a distance value between the interest vector of the user and the recommendation vector of each document to be recommended, and push the document to be recommended, which meets a preset condition, to the user as push information according to each distance value.
In an embodiment, the recommending unit 305 is specifically configured to recommend, as recommendation information, a document to be recommended of a recommendation vector corresponding to a preset number of minimum distance values among the plurality of distance values to the user.
It should be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the information recommendation apparatus 300 and each unit described above may refer to corresponding processes in the foregoing information recommendation method embodiments, and are not described herein again.
The information recommendation device 300 in this embodiment can recommend to the user by combining the browsing data of the user in the first preset time period and the second preset time period, so as to improve the accuracy and the rationality of information recommendation.
The information recommendation apparatus may be implemented in the form of a computer program that can be run on a computer device as shown in fig. 6. Referring to fig. 6, fig. 6 is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device 500 includes a processor 502, memory, and a network interface 505 connected by a system bus 501, where the memory may include a non-volatile storage medium 503 and an internal memory 504. The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer programs 5032 include program instructions that, when executed, cause the processor 502 to perform an information recommendation method. The processor 502 is used to provide computing and control capabilities that support the operation of the overall computer device 500. The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503, and when the computer program 5032 is executed by the processor 502, the processor 502 can be caused to execute an information recommendation method. The network interface 505 is used for network communication such as sending assigned tasks and the like. Those skilled in the art will appreciate that the configuration shown in fig. 6 is a block diagram of only a portion of the configuration associated with the present application and does not constitute a limitation of the computer device 500 to which the present application may be applied, and that a particular computer device 500 may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
Wherein the processor 502 is configured to run the computer program 5032 stored in the memory to implement the following functions: acquiring first browsing data of a plurality of users in a first preset time period and second browsing data of the users in a second preset time period, wherein the first browsing data and the second browsing data are user behavior data of the users when the users browse webpages; determining a plurality of user keywords and the interest degree of each user to each user keyword according to the first browsing data and the second browsing data, and generating an interest vector corresponding to each user according to the user keywords and the interest degrees of each user to each user keyword; acquiring a plurality of documents to be recommended, and acquiring document keywords corresponding to each document to be recommended and a weight value corresponding to each document keyword based on a preset keyword information technology; generating a recommendation vector corresponding to each document to be recommended according to the document keywords corresponding to the document to be recommended and the weight value corresponding to each document keyword; and calculating a distance value between the interest vector of the user and the recommendation vector of each document to be recommended, and pushing the document to be recommended meeting a preset condition to the user as push information according to each distance value.
In an embodiment, the first browsing data includes a plurality of documents browsed by the user in the first preset time period and a plurality of browsing behavior parameters of the user for each document in the first preset time period; the second browsing data comprises a plurality of documents browsed by the user in the second preset time period and a plurality of browsing behavior parameters of the user for each document in the second preset time period; when the processor 502 determines a plurality of user keywords and the interest level of each user for each user keyword according to the first browsing data and the second browsing data, the following functions are specifically implemented: based on a document theme generation model, acquiring a plurality of first themes corresponding to a plurality of documents in the first browsing data and a first keyword list corresponding to each first theme, and acquiring a plurality of second themes corresponding to a plurality of documents in the second browsing data and a second keyword list corresponding to each second theme, wherein the first keyword list and the second keyword list both include a plurality of theme keywords corresponding to corresponding themes and a weight value corresponding to each theme keyword; performing union operation on the plurality of topic keywords in the first browsing data and the plurality of topic keywords in the second browsing data to obtain a plurality of user keywords; based on a preset calculation rule, calculating the interest degree of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameters in the first browsing data, and calculating the interest degree of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameters in the second browsing data; acquiring a first weight value corresponding to the first preset time period and a second weight value corresponding to the second preset time period; and taking the first weight value as the weight of the interest degree of each topic keyword in the first browsing data and the second weight value as the weight of the interest degree of each topic keyword in the second browsing data, and calculating the interest degree of each user on each user keyword according to a preset calculation formula.
In an embodiment, when the processor 502 calculates, based on a preset calculation rule, a degree of interest of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameter in the first browsing data, and calculates a degree of interest of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameter in the second browsing data, the following functions are specifically implemented: calculating the interest degree of each user in each document in the first browsing data and the interest degree of each document in the second browsing data according to the browsing behavior parameters of each user, the word count of each document and the current browsing time; respectively acquiring browsing keywords of each document in the first browsing data and a weight value corresponding to each browsing keyword based on the preset keyword information technology, and acquiring browsing keywords of each document in the second browsing data and a weight value corresponding to each browsing keyword; calculating the probability of each document in the first browsing data on each first topic according to browsing keywords of each document in the first browsing data, a weight value corresponding to each browsing keyword, a plurality of first topics in the first browsing data and a first keyword list corresponding to each first topic; calculating the probability of each document in the second browsing data on each second topic according to the browsing keyword of each document in the second browsing data and the weight value corresponding to each browsing keyword, and a plurality of second topics in the second browsing data and a second keyword list corresponding to each second topic; obtaining the interest degree of each user in each first topic according to the interest degree of each user in each document in the first browsing data and the probability of each document in the first browsing data on each first topic; obtaining the interest degree of each user in each second topic according to the interest degree of each user in each document in the second browsing data and the probability of each document in the second browsing data on each second topic; calculating the interest degree of each user for each topic keyword in the first browsing data according to the interest degree of each user for each first topic, a plurality of topic keywords in a first keyword list of each first topic and a weight value corresponding to each topic keyword; and calculating the interest degree of each user for each topic keyword in the second browsing data according to the interest degree of each user for each second topic, a plurality of topic keywords in a second keyword list of each second topic and a weight value corresponding to each topic keyword.
In an embodiment, the browsing behavior parameters include a click parameter of each of the users on each of the documents in the first browsing data or the second browsing data, a start time and an end time of each of the users when browsing each of the documents; the processor 502 specifically implements the following functions when calculating the interest level of each user in each document in the first browsing data and the interest level of each document in the second browsing data according to the browsing behavior parameter of each user, the word count of each document, and the current browsing time: acquiring all documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document, and acquiring all documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document; counting the total word count and the total time consumed by each user in all the documents browsed by the user in the first preset time period according to all the documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document; counting the total word number and the total time consumed by all the documents browsed by each user in the second preset time period according to all the documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document; calculating a first browsing speed of each user according to the total word number and the total time of all the documents browsed by each user in the first preset time period; calculating a second browsing speed of each user according to the total word number and the total time consumed by all the documents browsed by each user in the second preset time period; calculating the attention degree of each user to each document browsed in the first preset time period according to the first browsing speed of each user, the starting time and the ending time of browsing each document in the first preset time period and the word number of each document; calculating the attention degree of each user to each document browsed in the second preset time period according to the second browsing speed of each user, the starting time and the ending time of browsing each document in the second preset time period and the word number of each document; calculating the interest degree of each user in each document in the first browsing data according to the attention degree of each user to each document browsed in the first preset time period, the click parameter of each user to each document in the first browsing data, the starting time of browsing each document and the current browsing time; and calculating the interest degree of each user in each document in the second browsing data according to the attention degree of each user to each document browsed in the second preset time period, the click parameter of each user to each document in the second browsing data, the starting time of browsing each document and the current browsing time.
In an embodiment, the processor 502 further implements the following functions before performing the counting of the total word count and the total time consumed by each user for browsing all the documents in the first preset time period according to all the documents browsed by each user in the first preset time period and the starting time and the ending time for browsing each document by each user: calculating the time length of each user to each document browsed in the first preset time period and calculating the time length of each user to each document browsed in the second preset time period according to the starting time and the ending time of each article browsed by the user; and respectively extracting the effective duration of each user in the first preset time period and the second preset time period and the document corresponding to the effective duration through normal distribution. Correspondingly, when the processor 502 performs statistics on the total word count and the total time consumed by each user for browsing all documents within the first preset time period according to all documents browsed by each user within the first preset time period and the start time and the end time for browsing each document, the following functions are specifically implemented: counting the total word number of the documents corresponding to all the effective durations of each user in the first preset time period and the total time corresponding to all the effective durations according to the documents corresponding to all the effective durations browsed by each user in the first preset time period and the corresponding effective durations; the processor 502 specifically implements the following functions when counting the total word count and the total time consumed by each user for browsing all documents within the second preset time period according to all documents browsed by each user within the second preset time period and the start time and the end time for browsing each document: and counting the total word number of the documents corresponding to all the effective durations of the users in the second preset time period and the total time corresponding to all the effective durations according to the documents corresponding to all the effective durations browsed by the users in the second preset time period and the corresponding effective durations.
In an embodiment, when the processor 502 executes to push the document to be recommended meeting the preset condition as push information to the user according to each distance value, the following functions are specifically implemented: and recommending the document to be recommended of the recommendation vector corresponding to the minimum distance value of the preset number of the distance values to the user as recommendation information.
It should be understood that, in the embodiment of the present application, the processor 502 may be a Central Processing Unit (CPU), and the processor 502 may also be other general processors, digital signal processors, application specific integrated circuits, ready-made programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It will be understood by those skilled in the art that all or part of the flow in the above-described embodiment of the information recommendation method may be implemented by a computer program instructing associated hardware. The computer program may be stored in a computer readable storage medium. The computer program is executed by at least one processor in the computer system to implement the process steps of the embodiments including the information recommendation methods described above.
The storage medium may be various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a magnetic disk, or an optical disk.
The steps in the method of the embodiment of the application can be sequentially adjusted, combined and deleted according to actual needs. The units in the device of the embodiment of the application can be combined, divided and deleted according to actual needs. The integrated unit, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in a storage medium. Based on such understanding, the technical solution of the present application may be substantially or partially implemented in the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a terminal, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application.
While the invention has been described with reference to specific embodiments, the scope of the invention is not limited thereto, and those skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (6)

1. An information recommendation method, comprising:
acquiring first browsing data of a plurality of users in a first preset time period and second browsing data of the users in a second preset time period, wherein the first browsing data and the second browsing data are user behavior data of the users when the users browse webpages;
determining a plurality of user keywords and the interest degree of each user to each user keyword according to the first browsing data and the second browsing data, and generating an interest vector corresponding to each user according to the user keywords and the interest degrees of each user to each user keyword;
acquiring a plurality of documents to be recommended, and acquiring document keywords corresponding to each document to be recommended and a weight value corresponding to each document keyword based on a preset keyword information technology;
generating a recommendation vector corresponding to each document to be recommended according to the document keywords corresponding to the document to be recommended and the weight value corresponding to each document keyword; and
calculating a distance value between the interest vector of the user and the recommendation vector of each document to be recommended, and pushing the document to be recommended meeting a preset condition to the user as push information according to each distance value;
the first browsing data comprises a plurality of documents browsed by the user within the first preset time period and a plurality of browsing behavior parameters of the user for each document within the first preset time period; the second browsing data comprises a plurality of documents browsed by the user in the second preset time period and a plurality of browsing behavior parameters of the user for each document in the second preset time period;
the determining a plurality of user keywords and the interest level of each user for each user keyword according to the first browsing data and the second browsing data comprises:
based on a document theme generation model, acquiring a plurality of first themes corresponding to a plurality of documents in the first browsing data and a first keyword list corresponding to each first theme, and acquiring a plurality of second themes corresponding to a plurality of documents in the second browsing data and a second keyword list corresponding to each second theme, wherein the first keyword list and the second keyword list both include a plurality of theme keywords corresponding to corresponding themes and a weight value corresponding to each theme keyword;
performing union operation on the plurality of topic keywords in the first browsing data and the plurality of topic keywords in the second browsing data to obtain a plurality of user keywords;
based on a preset calculation rule, calculating the interest degree of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameters in the first browsing data, and calculating the interest degree of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameters in the second browsing data;
acquiring a first weight value corresponding to the first preset time period and a second weight value corresponding to the second preset time period; and
taking the first weighted value as the weight of the interest degree of each topic keyword in the first browsing data and the second weighted value as the weight of the interest degree of each topic keyword in the second browsing data, and calculating the interest degree of each user keyword according to a preset calculation formula;
the method includes the steps of respectively calculating the interest degree of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameters in the first browsing data and calculating the interest degree of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameters in the second browsing data based on preset calculation rules, and includes the following steps:
calculating the interest degree of each user in each document in the first browsing data and the interest degree of each document in the second browsing data according to the browsing behavior parameters of each user, the word number of each document and the current browsing time;
respectively acquiring browsing keywords of each document in the first browsing data and a weight value corresponding to each browsing keyword based on the preset keyword information technology, and acquiring browsing keywords of each document in the second browsing data and a weight value corresponding to each browsing keyword;
calculating the probability of each document in the first browsing data on each first topic according to browsing keywords of each document in the first browsing data, a weight value corresponding to each browsing keyword, a plurality of first topics in the first browsing data and a first keyword list corresponding to each first topic;
calculating the probability of each document in the second browsing data on each second topic according to the browsing keyword of each document in the second browsing data and the weight value corresponding to each browsing keyword, and a plurality of second topics in the second browsing data and a second keyword list corresponding to each second topic;
obtaining the interest degree of each user in each first topic according to the interest degree of each user in each document in the first browsing data and the probability of each document in the first browsing data on each first topic;
obtaining the interest degree of each user in each second topic according to the interest degree of each user in each document in the second browsing data and the probability of each document in the second browsing data on each second topic;
calculating the interest degree of each user in each topic keyword in the first browsing data according to the interest degree of each user in each first topic, a plurality of topic keywords in a first keyword list of each first topic and a weight value corresponding to each topic keyword; and
calculating the interest degree of each user for each topic keyword in the second browsing data according to the interest degree of each user for each second topic, a plurality of topic keywords in a second keyword list of each second topic and a weight value corresponding to each topic keyword;
the browsing behavior parameters comprise click parameters of each user on each document in the first browsing data or the second browsing data, and start time and end time of each user when browsing each document;
the calculating, according to the browsing behavior parameter of each user, the word count of each document, and the current browsing time, a degree of interest of each user in each document in the first browsing data and a degree of interest of each document in the second browsing data, including:
acquiring all documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document, and acquiring all documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document;
counting the total word number and the total time consumed by all the documents browsed by each user in the first preset time period according to all the documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document;
counting the total word number and the total time consumed by all the documents browsed by each user in the second preset time period according to all the documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document;
calculating a first browsing speed of each user according to the total word number and the total time of all the documents browsed by each user in the first preset time period;
calculating a second browsing speed of each user according to the total word number and the total time consumed by all the documents browsed by each user in the second preset time period;
calculating the attention degree of each user to each document browsed in the first preset time period according to the first browsing speed of each user, the starting time and the ending time of browsing each document in the first preset time period and the word number of each document;
calculating the attention degree of each user to each document browsed in the second preset time period according to the second browsing speed of each user, the starting time and the ending time of browsing each document in the second preset time period and the word number of each document;
calculating the interest degree of each user in each document in the first browsing data according to the attention degree of each user to each document browsed in the first preset time period, the click parameter of each user to each document in the first browsing data, the starting time of browsing each document and the current browsing time; and
and calculating the interest degree of each user in each document in the second browsing data according to the attention degree of each user to each document browsed in the second preset time period, the click parameter of each user to each document in the second browsing data, the starting time of browsing each document and the current browsing time.
2. The information recommendation method according to claim 1, further comprising, before said counting total word counts and total time consumed for all documents browsed by each of said users in said first preset time period according to all documents browsed by each of said users in said first preset time period and start time and end time of browsing each document, a step of: calculating the time length of each user to each document browsed in the first preset time period and calculating the time length of each user to each document browsed in the second preset time period according to the starting time and the ending time of each article browsed by the user; extracting the effective duration of each user in the first preset time period and the second preset time period and the document corresponding to the effective duration respectively through normal distribution;
the counting total word count and total time consumed by each user in all the documents browsed by the user in the first preset time period according to all the documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document, comprising: counting the total word number of the documents corresponding to all the effective durations and the total time corresponding to all the effective durations of each user in the first preset time period according to the documents corresponding to all the effective durations browsed by each user in the first preset time period and the corresponding effective durations;
the counting total word count and total time consumed by each user in all the documents browsed by the user in the second preset time period according to all the documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document, includes: and counting the total word number of the documents corresponding to all the effective durations of the users in the second preset time period and the total time corresponding to all the effective durations according to the documents corresponding to all the effective durations browsed by the users in the second preset time period and the corresponding effective durations.
3. The information recommendation method according to claim 1, wherein the pushing the document to be recommended meeting the preset condition to the user as push information according to each distance value comprises: and recommending the document to be recommended of the recommendation vector corresponding to the minimum distance value of the preset number of the distance values to the user as recommendation information.
4. An information recommendation apparatus, comprising:
the browsing data acquisition unit is used for acquiring first browsing data of a plurality of users in a first preset time period and second browsing data of the users in a second preset time period, wherein the first browsing data and the second browsing data are user behavior data when the users browse webpages;
an interest vector generating unit, configured to determine, according to the first browsing data and the second browsing data, a plurality of user keywords and an interest level of each user for each user keyword, and generate an interest vector corresponding to each user according to the plurality of user keywords and the interest level of each user for each user keyword;
the system comprises a keyword acquisition unit, a recommendation unit and a recommendation unit, wherein the keyword acquisition unit is used for acquiring a plurality of documents to be recommended and acquiring document keywords corresponding to each document to be recommended and a weight value corresponding to each document keyword based on a preset keyword information technology;
the recommendation vector generation unit is used for generating a recommendation vector corresponding to each document to be recommended according to the document keywords corresponding to the document to be recommended and the weight value corresponding to each document keyword; and
the recommendation unit is used for calculating a distance value between the interest vector of the user and the recommendation vector of each document to be recommended, and pushing the document to be recommended meeting a preset condition to the user as push information according to each distance value;
the first browsing data comprises a plurality of documents browsed by the user within the first preset time period and a plurality of browsing behavior parameters of the user for each document within the first preset time period; the second browsing data comprises a plurality of documents browsed by the user within the second preset time period and a plurality of browsing behavior parameters of the user for each document within the second preset time period;
the interest vector generating unit is specifically configured to obtain, based on a document topic generation model, a plurality of first topics corresponding to a plurality of documents in the first browsing data and a first keyword list corresponding to each of the first topics, and obtain a plurality of second topics corresponding to a plurality of documents in the second browsing data and a second keyword list corresponding to each of the second topics, where the first keyword list and the second keyword list both include a plurality of topic keywords corresponding to corresponding topics and a weight value corresponding to each of the topic keywords; performing union operation on the plurality of topic keywords in the first browsing data and the plurality of topic keywords in the second browsing data to obtain a plurality of user keywords; based on a preset calculation rule, calculating the interest degree of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameters in the first browsing data, and calculating the interest degree of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameters in the second browsing data; acquiring a first weight value corresponding to the first preset time period and a second weight value corresponding to the second preset time period; the first weighted value is used as the weight of the interest degree of each topic keyword in the first browsing data, the second weighted value is used as the weight of the interest degree of each topic keyword in the second browsing data, and the interest degree of each user on each user keyword is calculated according to a preset calculation formula;
the method includes the steps of respectively calculating the interest degree of each user in each topic keyword in the first browsing data according to the document and the browsing behavior parameters in the first browsing data and calculating the interest degree of each user in each topic keyword in the second browsing data according to the document and the browsing behavior parameters in the second browsing data based on preset calculation rules, and includes the following steps:
calculating the interest degree of each user in each document in the first browsing data and the interest degree of each document in the second browsing data according to the browsing behavior parameters of each user, the word number of each document and the current browsing time;
respectively acquiring browsing keywords of each document in the first browsing data and a weight value corresponding to each browsing keyword based on the preset keyword information technology, and acquiring browsing keywords of each document in the second browsing data and a weight value corresponding to each browsing keyword;
calculating the probability of each document in the first browsing data on each first topic according to the browsing keyword of each document in the first browsing data and the weight value corresponding to each browsing keyword, and the plurality of first topics in the first browsing data and the first keyword list corresponding to each first topic;
calculating the probability of each document in the second browsing data on each second topic according to the browsing keyword of each document in the second browsing data and the weight value corresponding to each browsing keyword, and a plurality of second topics in the second browsing data and a second keyword list corresponding to each second topic;
obtaining the interest degree of each user in each first topic according to the interest degree of each user in each document in the first browsing data and the probability of each document in the first browsing data on each first topic;
obtaining the interest degree of each user in each second topic according to the interest degree of each user in each document in the second browsing data and the probability of each document in the second browsing data on each second topic;
calculating the interest degree of each user for each topic keyword in the first browsing data according to the interest degree of each user for each first topic, a plurality of topic keywords in a first keyword list of each first topic and a weight value corresponding to each topic keyword; and
calculating the interest degree of each user in each topic keyword in the second browsing data according to the interest degree of each user in each second topic, a plurality of topic keywords in a second keyword list of each second topic and a weight value corresponding to each topic keyword;
the browsing behavior parameters comprise click parameters of each user on each document in the first browsing data or the second browsing data, and start time and end time of each user when browsing each document;
the calculating, according to the browsing behavior parameter of each user, the word count of each document, and the current browsing time, a degree of interest of each user in each document in the first browsing data and a degree of interest of each document in the second browsing data, including:
acquiring all documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document, and acquiring all documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document;
counting the total word count and the total time consumed by each user in all the documents browsed by the user in the first preset time period according to all the documents browsed by each user in the first preset time period and the starting time and the ending time of browsing each document;
counting the total word number and the total time consumed by all the documents browsed by each user in the second preset time period according to all the documents browsed by each user in the second preset time period and the starting time and the ending time of browsing each document;
calculating a first browsing speed of each user according to the total word number and the total time of all the documents browsed by each user in the first preset time period;
calculating a second browsing speed of each user according to the total word number and the total time consumed by all the documents browsed by each user in the second preset time period;
calculating the attention degree of each user to each document browsed in the first preset time period according to the first browsing speed of each user, the starting time and the ending time of browsing each document in the first preset time period and the word number of each document;
calculating the attention degree of each user to each document browsed in the second preset time period according to the second browsing speed of each user, the starting time and the ending time of browsing each document in the second preset time period and the word number of each document;
calculating the interest degree of each user in each document in the first browsing data according to the attention degree of each user to each document browsed in the first preset time period, the click parameter of each user to each document in the first browsing data, the starting time of browsing each document and the current browsing time; and
and calculating the interest degree of each user in each document in the second browsing data according to the attention degree of each user to each document browsed in the second preset time period, the click parameter of each user to each document in the second browsing data, the starting time of browsing each document and the current browsing time.
5. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the information recommendation method according to any one of claims 1 to 3 when executing the computer program.
6. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a processor, causes the processor to execute the information recommendation method according to any one of claims 1 to 3.
CN201810948488.9A 2018-08-20 2018-08-20 Information recommendation method and device, computer equipment and storage medium Active CN109190024B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201810948488.9A CN109190024B (en) 2018-08-20 2018-08-20 Information recommendation method and device, computer equipment and storage medium
PCT/CN2018/125327 WO2020037930A1 (en) 2018-08-20 2018-12-29 Method, device, and computer device for recommending information, and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810948488.9A CN109190024B (en) 2018-08-20 2018-08-20 Information recommendation method and device, computer equipment and storage medium

Publications (2)

Publication Number Publication Date
CN109190024A CN109190024A (en) 2019-01-11
CN109190024B true CN109190024B (en) 2023-04-07

Family

ID=64918962

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810948488.9A Active CN109190024B (en) 2018-08-20 2018-08-20 Information recommendation method and device, computer equipment and storage medium

Country Status (2)

Country Link
CN (1) CN109190024B (en)
WO (1) WO2020037930A1 (en)

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110008403B (en) * 2019-03-05 2021-05-28 百度在线网络技术（北京）有限公司 Target information sorting method, sorting system, recommendation method and recommendation system
CN111046222B (en) * 2019-04-29 2023-06-27 广东小天才科技有限公司 Dictation list generation method and electronic equipment
CN110347900B (en) * 2019-07-10 2022-12-27 腾讯科技（深圳）有限公司 Keyword importance calculation method, device, server and medium
CN112995248B (en) * 2019-12-12 2023-04-07 阿里巴巴集团控股有限公司 Information pushing method, device and equipment
CN111143689A (en) * 2019-12-31 2020-05-12 青梧桐有限责任公司 Method for constructing recommendation engine according to user requirements and user portrait
US20210357983A1 (en) * 2020-05-14 2021-11-18 Nanning Fugui Precision Industrial Co., Ltd. System for presenting advertisements online and method thereof
CN113676505B (en) * 2020-05-15 2023-11-28 财付通支付科技有限公司 Information pushing method, device, computer equipment and storage medium
CN113689019A (en) * 2020-05-18 2021-11-23 佛山市顺德区美的电热电器制造有限公司 Data processing method, device and system
CN112749331B (en) * 2020-06-28 2023-09-19 腾讯科技（深圳）有限公司 Data recommendation method and device and computer readable storage medium
CN112328881B (en) * 2020-11-05 2024-04-02 中国平安人寿保险股份有限公司 Article recommendation method, device, terminal equipment and storage medium
CN112364155B (en) * 2020-11-20 2024-05-31 北京五八信息技术有限公司 Information processing method and device
CN113051480A (en) * 2021-04-22 2021-06-29 深圳壹账通智能科技有限公司 Resource pushing method and device, electronic equipment and storage medium
CN112992154A (en) * 2021-05-08 2021-06-18 北京远鉴信息技术有限公司 Voice identity determination method and system based on enhanced voiceprint library
CN112989824A (en) * 2021-05-12 2021-06-18 武汉卓尔数字传媒科技有限公司 Information pushing method and device, electronic equipment and storage medium
CN113360753A (en) * 2021-05-26 2021-09-07 平安国际智慧城市科技股份有限公司 Information recommendation method, device, equipment and medium based on user historical behaviors
CN113553509B (en) * 2021-07-29 2024-03-01 北京达佳互联信息技术有限公司 Content recommendation method and device, electronic equipment and storage medium
CN113780415B (en) * 2021-09-10 2023-08-15 平安科技（深圳）有限公司 User portrait generating method, device, equipment and medium based on applet game
CN114398547B (en) * 2022-01-06 2022-09-30 北京博瑞彤芸科技股份有限公司 Processing method and device for intelligently pushing articles
CN115577167B (en) * 2022-08-29 2023-11-21 建信金融科技有限责任公司 Webassembl-based content recommendation method, device, equipment and storage medium
CN116089624B (en) * 2022-11-17 2024-02-27 昆仑数智科技有限责任公司 Knowledge graph-based data recommendation method, device and system
CN117575745B (en) * 2024-01-17 2024-04-30 山东正禾大教育科技有限公司 Course teaching resource individual recommendation method based on AI big data
CN117648462B (en) * 2024-01-29 2024-06-18 深圳感臻智能股份有限公司 Video recommendation method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103235824A (en) * 2013-05-06 2013-08-07 上海河广信息科技有限公司 Method and system for determining web page texts users interested in according to browsed web pages
CN103714084A (en) * 2012-10-08 2014-04-09 腾讯科技（深圳）有限公司 Method and device for recommending information
CN104199874A (en) * 2014-08-20 2014-12-10 哈尔滨工程大学 Webpage recommendation method based on user browsing behaviors
CN104899273A (en) * 2015-05-27 2015-09-09 东南大学 Personalized webpage recommendation method based on topic and relative entropy
CN108280114A (en) * 2017-07-28 2018-07-13 淮阴工学院 A kind of user's literature reading interest analysis method based on deep learning

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103729360A (en) * 2012-10-12 2014-04-16 腾讯科技（深圳）有限公司 Interest label recommendation method and system
US20150262069A1 (en) * 2014-03-11 2015-09-17 Delvv, Inc. Automatic topic and interest based content recommendation system for mobile devices
CN106055661B (en) * 2016-06-02 2017-11-17 福州大学 More interest resource recommendations based on more Markov chain models
CN106407418A (en) * 2016-09-23 2017-02-15 Tcl集团股份有限公司 A face identification-based personalized video recommendation method and recommendation system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103714084A (en) * 2012-10-08 2014-04-09 腾讯科技（深圳）有限公司 Method and device for recommending information
CN103235824A (en) * 2013-05-06 2013-08-07 上海河广信息科技有限公司 Method and system for determining web page texts users interested in according to browsed web pages
CN104199874A (en) * 2014-08-20 2014-12-10 哈尔滨工程大学 Webpage recommendation method based on user browsing behaviors
CN104899273A (en) * 2015-05-27 2015-09-09 东南大学 Personalized webpage recommendation method based on topic and relative entropy
CN108280114A (en) * 2017-07-28 2018-07-13 淮阴工学院 A kind of user's literature reading interest analysis method based on deep learning

Also Published As

Publication number Publication date
WO2020037930A1 (en) 2020-02-27
CN109190024A (en) 2019-01-11

Similar Documents

Publication Publication Date Title
CN109190024B (en) Information recommendation method and device, computer equipment and storage medium
US11710054B2 (en) Information recommendation method, apparatus, and server based on user data in an online forum
JP5717858B2 (en) Text set matching
WO2020048084A1 (en) Resource recommendation method and apparatus, computer device, and computer-readable storage medium
TWI582619B (en) Method and apparatus for providing referral words
JP5891213B2 (en) A posteriori probability calculation device, posterior probability calculation method, and program
US10169449B2 (en) Method, apparatus, and server for acquiring recommended topic
CN107784066A (en) Information recommendation method, device, server and storage medium
CN106649681B (en) Data processing method, device and equipment
US10346496B2 (en) Information category obtaining method and apparatus
CN105868256A (en) Method and system for processing user behavior data
CN105930507B (en) A kind of method and device for the web browsing interest obtaining user
CN105528416B (en) A kind of monitoring method and system of network upgrade content
CN111259220B (en) Data acquisition method and system based on big data
JP2011227721A (en) Interest extraction device, interest extraction method, and interest extraction program
WO2013110357A1 (en) Social network analysis
CN111932308A (en) Data recommendation method, device and equipment
CN111651666A (en) User theme recommendation method and device, computer equipment and storage medium
US20140280216A1 (en) Automated ranking of contributors to a knowledge base
CN115687810A (en) Webpage searching method and device and related equipment
CN108228598B (en) Media information sorting method, server and system
CN115203565A (en) Cold start method and device of recommendation system, electronic equipment and storage medium
CN109828970B (en) Information processing method and device and electronic equipment
CN113868373A (en) Word cloud generation method and device, electronic equipment and storage medium
CN113392329A (en) Content recommendation method and device, electronic equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant