CN104035961A - Method and system for recognizing social internet population - Google Patents

Method and system for recognizing social internet population Download PDF

Info

Publication number
CN104035961A
CN104035961A CN201410194361.4A CN201410194361A CN104035961A CN 104035961 A CN104035961 A CN 104035961A CN 201410194361 A CN201410194361 A CN 201410194361A CN 104035961 A CN104035961 A CN 104035961A
Authority
CN
China
Prior art keywords
user
message
theme
identified
probability
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.)
Granted
Application number
CN201410194361.4A
Other languages
Chinese (zh)
Other versions
CN104035961B (en
Inventor
怀进鹏
武南南
李建欣
张日崇
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.)
Beihang University
Original Assignee
Beihang University
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 Beihang University filed Critical Beihang University
Priority to CN201410194361.4A priority Critical patent/CN104035961B/en
Publication of CN104035961A publication Critical patent/CN104035961A/en
Application granted granted Critical
Publication of CN104035961B publication Critical patent/CN104035961B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Primary Health Care (AREA)
  • Marketing (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Health & Medical Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The invention relates to a method and system for recognizing a social internet population. The method comprises the steps that a plurality of users to be recognized and user information corresponding to each user to be recognized in a social network are acquired; the user information comprises information content and infection states; the theme which each piece of user information belongs to and the probability that each piece of user information corresponds to the corresponding theme are determined according to the information content in the user information corresponding to each user to be recognized; for each user to be recognized, the probability that each theme is infected by the users to be recognized is calculated according to the probability that each piece of user information belongs to the corresponding theme and the infection state in each piece of user information, and therefore a user population corresponding to each theme can be determined according to the probability that each theme is infected by the users to be recognized, the range, the path and the influence causing information spreading can be predicated, and information spreading can be effectively controlled.

Description

The recognition methods of social networks colony and system
Technical field
The present invention relates to communication technical field, relate in particular to a kind of social networks colony's recognition methods and system.
Background technology
In prior art, in web1.0 period, in network, transmission of news pattern is mainly: publisher edits and gives out information, and masses browse message.And interacting activity between user and user seldom, therefore, the scope of message propagation, path and impact can Accurate Predictions.
Yet, in web2.0 period, development along with social networks such as microblogging, QQ, Renren Network, Face book, interacting activity between user and user rolls up, each user can become transmission of news source, cause scope, path and the impact of message propagation to be difficult to prediction, be difficult to identify the key user colony that makes message wide-scale distribution from social networks, thereby be difficult to transmission of news effectively to be controlled.
Summary of the invention
The invention provides a kind of social networks colony's recognition methods and system, for solving prior art, be difficult to identify the problem of the key user colony that makes message wide-scale distribution from social networks.
First aspect of the present invention is to provide the recognition methods of a kind of social networks colony, comprising:
Obtain a plurality of users to be identified and user message corresponding to each user to be identified in social networks; Described user message comprises message content and Infection Status;
For each user to be identified, according to the message content in user message corresponding to each user to be identified, the probability that the theme described in determining under each user message and described each user message belong to corresponding theme;
For each user to be identified, according to described each user message, belong to the probability of corresponding theme, and the Infection Status in described each user message, calculate the probability that each theme is infected by described user to be identified;
For each user to be identified, the probability being infected by described user to be identified according to each theme, determines the user group corresponding with theme that described user to be identified is affiliated.
Another aspect of the present invention provides a kind of social networks colony recognition system, comprising:
Acquisition module, for obtaining a plurality of users to be identified of social networks and user message corresponding to each user to be identified; Described user message comprises message content and Infection Status;
Determination module, for for each user to be identified, according to the message content in user message corresponding to each user to be identified, the probability that the theme described in determining under each user message and described each user message belong to corresponding theme;
Computing module, for for each user to be identified, belongs to the probability of corresponding theme according to described each user message, and the Infection Status in described each user message, calculates the probability that each theme is infected by described user to be identified;
Described determination module, also, for for each user to be identified, by the probability of described user to be identified infection, determines the user group corresponding with theme that described user to be identified is affiliated according to each theme.
In the present invention, by obtaining a plurality of users to be identified and user message corresponding to each user to be identified in social networks; User message comprises message content and Infection Status; For each user to be identified, according to the message content in user message corresponding to each user to be identified, determine the probability that theme under each user message and each user message belong to corresponding theme; For each user to be identified, the probability that belongs to corresponding theme according to each user message, and the Infection Status in each user message, calculate the probability that each theme is infected by user to be identified, thereby the probability being infected by user to be identified according to each theme, can determine the user group that each theme is corresponding, can be on causing scope, path and the impact of message propagation to be predicted, and then transmission of news is effectively controlled.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of an embodiment of social networks colony's recognition methods provided by the invention;
Fig. 2 is the process flow diagram of social networks another embodiment of recognition methods of colony provided by the invention;
Fig. 3 is the schematic diagram of relational network;
Fig. 4 is the structural representation of an embodiment of social networks colony's recognition system provided by the invention;
Fig. 5 is the structural representation of social networks another embodiment of recognition system of colony provided by the invention.
Embodiment
For making object, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.
Fig. 1 is the process flow diagram of an embodiment of social networks colony's recognition methods provided by the invention, as shown in Figure 1, comprising:
101, obtain a plurality of users to be identified and user message corresponding to each user to be identified in social networks, user message comprises message content and Infection Status.
The executive agent of social networks provided by the invention colony recognition methods can be social networks colony recognition system, and social networks colony recognition system is specially the Software tool being arranged on social networks server.
Because the number of users in social networks is very big, the quantity of user message is also very big, social networks colony recognition system is difficult to obtain user message corresponding to all users to be analyzed, and in social networks, the larger user of impact is generally bean vermicelli number, is concerned number and the more user of message transmission times.Therefore, social networks colony recognition system can be according to user's bean vermicelli number, be concerned number or message transmission times etc. is selected user to be identified.For example, social networks colony recognition system can select user that bean vermicelli number is greater than default bean vermicelli number as user to be identified, or, select to be concerned number and be greater than and be defaultly concerned several users as user to be identified.
Social networks colony recognition system is obtained after user to be identified, the application programming interface (Application Programming Interface, API) that can utilize reptile instrument in social networks or social network sites to provide is obtained user and is reprinted, sends and all message of comment.All message of reprinting, sending and comment in conjunction with all users to be identified, obtain user message corresponding to each user to be identified.The message that user sends can comprise that user sends to other users' message and the message that user is transmitted to other users.For example, when user to be identified comprises user A, user B and user C, and user A reprints, sends or the message of comment comprises: message a and message b; User B reprints, sends or the message of comment comprises: message c and message d; User C reprints, sends or the message of comment comprises: message e.The user message that now user A is corresponding comprises: message a, message b, message c, message d and message e.Wherein, Infection Status refers to user and whether reprints, sends and commented on described message, if user reprints, sends or commented on described message, Infection Status is 1; If user does not reprint, do not send and do not comment on described message, Infection Status is 0.For example, the message that message a and message b reprint, send or comment on for user A, so the Infection Status in message a and message b is 1; Message c, message d and message e are the message that user B or user C reprint, send or comment on, so the Infection Status in message c, message d and message e is 0.
Wherein, message content refers to the particular content of the described message of user's reprinting or transmission.If the message that described message is user comment, message content is except comprising the particular content of described message, also comprises the comment of before comment that user carries out described message and other users, described message being carried out.
102,, for each user to be identified, according to the message content in user message corresponding to each user to be identified, determine the probability that theme under each user message and each user message belong to corresponding theme.
Particularly, social networks colony recognition system can be obtained all vocabulary in message content, according to the vocabulary in message content, determines the theme under message.Wherein, theme specifically can refer to the general idea of message content, and for example, theme can be: finance and economics, politics, physical culture, science and geography etc., the value volume and range of product of theme can be set in advance by social networks colony recognition system.When the general idea of message content relates to two or more themes, can determine that message belongs to the probability of corresponding theme according to the vocabulary quantity that in message content, each theme is corresponding.
103, for each user to be identified, according to each user message, belong to the probability of corresponding theme, and the Infection Status in each user message, calculate the probability that each theme is infected by user to be identified.
104,, for each user to be identified, the probability being infected by user to be identified according to each theme, determines the user group corresponding with theme that user to be identified is affiliated.
Particularly, suppose that the theme that user message corresponding to user relates to has 3 kinds, be respectively: finance and economics, politics and physical culture, the probability being infected by user to be identified according to each theme, wherein a kind of optional mode of determining the user group corresponding with theme that user to be identified is affiliated is: the default finance and economics threshold value if the probability that theme " finance and economics " is infected by user overflows, determine that described user belongs to the user group corresponding with " finance and economics "; The default political threshold value if the probability that theme " politics " is infected by user overflows, determines that described user belongs to the user group corresponding with " politics "; The default physical culture threshold value if the probability that theme " physical culture " is infected by user overflows, determines that described user belongs to the user group corresponding with " physical culture ".
Further, for each user to be identified, the probability being infected by user to be identified according to each theme, after determining the affiliated user group corresponding with theme of user to be identified, also comprises:
Obtain the message content of message to be analyzed; According to the message content of message to be analyzed, determine the theme that message to be analyzed is affiliated and the probability that belongs to corresponding theme; Obtain the theme of probable value maximum in the affiliated theme of message to be analyzed; By user group corresponding to the theme with probable value maximum, the user group who is defined as propagating message to be analyzed, to control transmission of news to be analyzed.
Wherein, message to be analyzed can be the popular message of wide-scale distribution.
In the present embodiment, by obtaining a plurality of users to be identified and user message corresponding to each user to be identified in social networks; User message comprises message content and Infection Status; For each user to be identified, according to the message content in user message corresponding to each user to be identified, determine the probability that theme under each user message and each user message belong to corresponding theme; For each user to be identified, the probability that belongs to corresponding theme according to each user message, and the Infection Status in each user message, calculate the probability that each theme is infected by user to be identified, thereby the probability being infected by user to be identified according to each theme, can determine the user group that each theme is corresponding, can be on causing scope, path and the impact of message propagation to be predicted, and then transmission of news is effectively controlled.
Fig. 2 is the process flow diagram of social networks another embodiment of recognition methods of colony provided by the invention, and as shown in Figure 2, on basis embodiment illustrated in fig. 1, step 102 specifically can comprise:
1021, for each user to be identified, the message content in user message is carried out to participle, obtain keyword in user message and the word frequency of keyword.
Wherein, the process of message content being carried out to participle refers to the process of obtaining all vocabulary in message content, obtains after all vocabulary in message content, deletes the vocabulary without concrete meaning in vocabulary, such as stop words etc., obtains the keyword in user message; The occurrence number of each keyword in counting user message, obtains the word frequency of keyword.
Need to describe, when the default vocabulary of social networks colony recognition system, the expression mode of keyword word frequency can be as shown in table 1.Wherein, n is the total degree that in message content, keyword occurs, word n represents that keyword is n word in vocabulary, and count n represents the number of times that keyword occurs in described message.
Table 1
n?word1:count1word2:count2…word?n:count?n
1022, the vocabulary default according to keyword query, determines the theme that user message is affiliated.
In default vocabulary, preserve all keywords in user message corresponding to user to be identified and the corresponding relation of keyword and theme.For example, basketball, football or in corresponding the theming as of superfine keyword " physical culture "; Corresponding the theming as of keyword " finance and economics " such as stock, real estate, economy, upper card.
1023,, according to the word frequency of keyword, determine that user message belongs to the probability of corresponding theme.
Wherein, the theme " finance and economics " of take describes as example, and the probability that user message belongs to theme " finance and economics " can be: the ratio that belongs to the total degree n that in the summation of keyword occurrence number of theme " finance and economics " and message content, each keyword occurs in user message.
Further, on Fig. 1 or basis embodiment illustrated in fig. 2, step 103 specifically can comprise:
For each user to be identified, for each theme, according to each user message, belong to the probability of corresponding theme, and the Infection Status in each user message, calculate the probability that in each user message, theme is infected by user to be identified; The probability being infected by user to be identified according to theme in each user message, determines the probability that theme is infected by user to be identified.
Below describe for example, if user's user message has 5, theme has 3 kinds: finance and economics, politics and physical culture, and the probability that article one message belongs to 3 themes is respectively 0.7,0.2, and 0.1; The probability that second message belongs to 3 themes is respectively 0.6,0.2, and 0.2; Article three, the probability that message belongs to 3 themes is respectively 0.9,0.0,0.1; Article four, the probability that message belongs to 3 themes is respectively 0.5,0.5,0.0; Article five, the probability that message belongs to 3 themes is respectively 1.0,0.0,0.0.And user is respectively the Infection Status of above-mentioned 5 message: 1,1,1,0 and 1, the probability that theme " finance and economics " is infected by user is (0.7*1+0.6*1+0.9*1+0.5*0+1.0*1)/5=0.64, the probability that theme " politics " is infected by user is (0.2*1+0.2*1+0.0*1+0.5*0+0.0*1)/5=0.08, and the probability that theme " physical culture " is infected by user is (0.1*1+0.2*1+0.1*1+0.0*0+0.0*1)/5=0.08.
What need to describe is, the accuracy of " probability that each theme is infected by user " that the account form in above-described embodiment calculates is limited, therefore in reality, " probability that each theme is infected by user " that social networks colony recognition system calculates afterwards, can verify " probability that each theme is infected by user ", the probability being infected by user according to each theme, in conjunction with the prior distribution in mathematics, the means such as posteriority distribution and likelihood function, whether the accuracy of verifying " probability that each theme is infected by user " that calculate meets preset requirement, and to " probability that each theme is infected by the user " adjustment that circulates.
Further, on the basis of above-described embodiment, step 104 specifically can comprise:
For each user to be identified, perpetual object, buddy list and the message of obtaining user to be identified send object, build relational network; For each theme, the probability being infected by user to be identified according to theme, determines the connection distance between any two users to be identified in relational network with annexation; If the connection distance having between two users to be identified of annexation is less than preset value, determine that two users to be identified belong to the user group that theme is corresponding.
Wherein, perpetual object, buddy list and the message of obtaining user to be identified send after object, social networks colony recognition system can send object according to user's to be identified perpetual object, buddy list and message and determine the annexation between user and other users, according to the annexation opening relationships network between user and other users.For example, when user to be identified comprises user A, B, C, D and E, the schematic diagram of relational network can be as shown in Figure 3.Optionally, for example, if in the unit interval (for example, every day or monthly in) user A number from message to user B that send is for example, while being less than default number (6), can delete the annexation between user A and user B.
Particularly, suppose for theme M, the computing formula in relational network with the connection distance between any two users of annexation is specifically as follows L=(1-P (A))+(1-P (B)), wherein, between the user A that L represents to have annexation and user B, be connected distance, P (A) represents the probability that the message of theme M is infected by user A, and P (B) represents the probability that the message of theme M is infected by user B.
In the present embodiment, by obtaining a plurality of users to be identified and user message corresponding to each user to be identified in social networks, user message comprises message content and Infection Status, for each user to be identified, message content in user message is carried out to participle, obtain keyword in user message and the word frequency of keyword, the vocabulary default according to keyword query, determine the theme that user message is affiliated, according to the word frequency of keyword, determine that user message belongs to the probability of corresponding theme, for each user to be identified, the probability that belongs to corresponding theme according to each user message, and the Infection Status in each user message, calculate the probability that each theme is infected by user to be identified, thereby the probability that social networks colony recognition system is infected by user according to each theme, can determine the user group that each theme is corresponding, can be to causing the scope of message propagation, path and impact are predicted, and then transmission of news is effectively controlled.
One of ordinary skill in the art will appreciate that: all or part of step that realizes above-mentioned each embodiment of the method can complete by the relevant hardware of programmed instruction.Aforesaid program can be stored in a computer read/write memory medium.This program, when carrying out, is carried out the step that comprises above-mentioned each embodiment of the method; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CDs.
Fig. 4 is the structural representation of an embodiment of social networks colony's recognition system provided by the invention, as shown in Figure 4, comprising:
Acquisition module 41, for obtaining a plurality of users to be identified of social networks and user message corresponding to each user to be identified, user message comprises message content and Infection Status.
Wherein, message content refers to the particular content of the described message of user's reprinting or transmission.If the message that described message is user comment, message content is except comprising the particular content of described message, also comprises the comment of before comment that user carries out described message and other users, described message being carried out.User refers to user to the Infection Status of message and whether reprints, sends and commented on described message, if user reprints, sends or commented on described message, user is 1 to the Infection Status of message; If user does not reprint, do not send and do not comment on described message, user is 0 to the Infection Status of message.
Determination module 42, for for each user to be identified, according to the message content in user message corresponding to each user to be identified, determines the probability that theme under each user message and each user message belong to corresponding theme.
Computing module 43, for for each user to be identified, belongs to the probability of corresponding theme according to each user message, and the Infection Status in each user message, calculates the probability that each theme is infected by user to be identified.
Determination module 42, also, for for each user to be identified, by the probability of user's infection to be identified, determines the user group corresponding with theme that user to be identified is affiliated according to each theme.
Further, acquisition module 41 also for, obtain the message content of message to be analyzed;
Determination module 42 also for, according to the message content of message to be analyzed, determine the theme under message to be analyzed and the probability that belongs to corresponding theme.
Acquisition module 41 also for, obtain the theme of probable value maximum in the theme under message to be analyzed.
Determination module 42 also for, by user group corresponding to the theme with probable value maximum, the user group who is defined as propagating message to be analyzed, to control transmission of news to be analyzed.
In the present embodiment, by obtaining a plurality of users to be identified and user message corresponding to each user to be identified in social networks; User message comprises message content and Infection Status; For each user to be identified, according to the message content in user message corresponding to each user to be identified, determine the probability that theme under each user message and each user message belong to corresponding theme; For each user to be identified, the probability that belongs to corresponding theme according to each user message, and the Infection Status in each user message, calculate the probability that each theme is infected by user to be identified, thereby the probability being infected by user to be identified according to each theme, can determine the user group that each theme is corresponding, can be on causing scope, path and the impact of message propagation to be predicted, and then transmission of news is effectively controlled.
Fig. 5 is the structural representation of social networks another embodiment of recognition system of colony provided by the invention, and as shown in Figure 5, on basis embodiment illustrated in fig. 4, determination module 42 can comprise:
Participle unit 421, for for each user to be identified, carries out participle to the message content in user message, obtains keyword in user message and the word frequency of keyword.
Wherein, the process of message content being carried out to participle refers to the process of obtaining all vocabulary in message content, obtains after all vocabulary in message content, deletes the vocabulary without concrete meaning in vocabulary, such as stop words etc., obtains the keyword in message; The occurrence number of each keyword in statistical message, obtains the word frequency of keyword.
Query unit 422, for the vocabulary default according to keyword query, determines the theme that user message is affiliated;
Determining unit 423, for according to the word frequency of keyword, determines that user message belongs to the probability of corresponding theme.
Further, on Fig. 4 or basis embodiment illustrated in fig. 5, computing module 43 specifically for,
For each user to be identified, for each theme, according to each user message, belong to the probability of corresponding theme, and the Infection Status in each user message, calculate the probability that in each user message, theme is infected by user to be identified; The probability being infected by user to be identified according to theme in each user message, determines the probability that in user message, theme is infected by user to be identified.
Further, determination module 42 also comprises:
Construction unit, for for each user to be identified, obtains user's to be identified perpetual object, buddy list and message transmission object, builds relational network; Determining unit 423, also, for for each theme, by the probability of user's infection to be identified, determines the connection distance between any two users to be identified in relational network with annexation according to theme; Determining unit 423, if be also less than preset value for the connection distance having between two users to be identified of annexation, determines that two users to be identified belong to the user group that theme is corresponding.
In the present embodiment, by obtaining a plurality of users to be identified and user message corresponding to each user to be identified in social networks, user message comprises message content and Infection Status, for each user to be identified, message content in user message is carried out to participle, obtain keyword in user message and the word frequency of keyword, the vocabulary default according to keyword query, determine the theme that user message is affiliated, according to the word frequency of keyword, determine that user message belongs to the probability of corresponding theme, for each user to be identified, the probability that belongs to corresponding theme according to each user message, and the Infection Status in each user message, calculate the probability that each theme is infected by user to be identified, thereby the probability that social networks colony recognition system is infected by user according to each theme, can determine the user group that each theme is corresponding, can be to causing the scope of message propagation, path and impact are predicted, and then transmission of news is effectively controlled.
Finally it should be noted that: each embodiment, only in order to technical scheme of the present invention to be described, is not intended to limit above; Although the present invention is had been described in detail with reference to aforementioned each embodiment, those of ordinary skill in the art is to be understood that: its technical scheme that still can record aforementioned each embodiment is modified, or some or all of technical characterictic is wherein equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.

Claims (10)

  1. The recognition methods of 1.Yi Zhong social networks colony, is characterized in that, comprising:
    Obtain a plurality of users to be identified and user message corresponding to each user to be identified in social networks; Described user message comprises message content and Infection Status;
    For each user to be identified, according to the message content in user message corresponding to each user to be identified, the probability that the theme described in determining under each user message and described each user message belong to corresponding theme;
    For each user to be identified, according to described each user message, belong to the probability of corresponding theme, and the Infection Status in described each user message, calculate the probability that each theme is infected by described user to be identified;
    For each user to be identified, the probability being infected by described user to be identified according to each theme, determines the user group corresponding with theme that described user to be identified is affiliated.
  2. 2. method according to claim 1, it is characterized in that, described for each user to be identified, according to the message content in user message corresponding to each user to be identified, the probability that theme described in determining under each user message and described each user message belong to corresponding theme, comprising:
    For each user to be identified, the message content in described user message is carried out to participle, obtain keyword in described user message and the word frequency of described keyword;
    The vocabulary default according to described keyword query, determines the theme that described user message is affiliated;
    According to the word frequency of described keyword, determine that described user message belongs to the probability of corresponding theme.
  3. 3. method according to claim 1 and 2, it is characterized in that, described for each user to be identified, the probability that belongs to corresponding theme according to described each user message, and the Infection Status in described each user message, calculate the message of each theme by the probability of described user to be identified infection, comprising:
    For each user to be identified, for each theme, the probability that belongs to corresponding theme according to described each user message, and the Infection Status in described each user message, calculate the probability that described in described each user message, theme is infected by described user to be identified;
    The probability being infected by described user to be identified according to theme described in described each user message, determines the probability that described theme is infected by described user to be identified.
  4. 4. method according to claim 3, is characterized in that, described for each user to be identified, and the probability being infected by described user to be identified according to each theme is determined the user group corresponding with theme that described user to be identified is affiliated, comprising:
    For each user to be identified, perpetual object, buddy list and the message of obtaining described user to be identified send object, build relational network;
    For each theme, the probability being infected by described user to be identified according to described theme, determines the connection distance between any two users to be identified in described relational network with annexation;
    If the connection distance having between two users to be identified of annexation is less than preset value, determine that described two users to be identified belong to the user group that described theme is corresponding.
  5. 5. method according to claim 1, it is characterized in that, described for each user to be identified, the probability being infected by described user to be identified according to each theme, after determining the affiliated user group corresponding with theme of described user to be identified, also comprise:
    Obtain the message content of message to be analyzed;
    According to the message content of described message to be analyzed, determine the theme that described message to be analyzed is affiliated and the probability that belongs to corresponding theme;
    Obtain the theme of probable value maximum in the affiliated theme of described message to be analyzed;
    By user group corresponding to the theme with described probable value maximum, the user group who is defined as propagating described message to be analyzed, to control described transmission of news to be analyzed.
  6. 6.Yi Zhong social networks colony recognition system, is characterized in that, comprising:
    Acquisition module, for obtaining a plurality of users to be identified of social networks and user message corresponding to each user to be identified; Described user message comprises message content and Infection Status;
    Determination module, for for each user to be identified, according to the message content in user message corresponding to each user to be identified, the probability that the theme described in determining under each user message and described each user message belong to corresponding theme;
    Computing module, for for each user to be identified, belongs to the probability of corresponding theme according to described each user message, and the Infection Status in described each user message, calculates the probability that each theme is infected by described user to be identified;
    Described determination module, also, for for each user to be identified, by the probability of described user to be identified infection, determines the user group corresponding with theme that described user to be identified is affiliated according to each theme.
  7. 7. system according to claim 6, is characterized in that, described determination module also comprises:
    Participle unit, for for each user to be identified, carries out participle to the message content in described user message, obtains keyword in described user message and the word frequency of described keyword;
    Query unit, for the vocabulary default according to described keyword query, determines the theme that described user message is affiliated;
    Determining unit, for according to the word frequency of described keyword, determines that described user message belongs to the probability of corresponding theme.
  8. 8. according to the system described in claim 6 or 7, it is characterized in that, described computing module specifically for,
    For each user to be identified, for each theme, the probability that belongs to corresponding theme according to described each user message, and the Infection Status in described each user message, calculate the probability that described in described each user message, theme is infected by described user to be identified;
    The probability being infected by described user to be identified according to theme described in described each user message, determines the probability that described theme is infected by described user to be identified.
  9. 9. system according to claim 8, is characterized in that, described determination module also comprises:
    Construction unit, for for each user to be identified, obtains described user's to be identified perpetual object, buddy list and message transmission object, builds relational network;
    Determining unit, also, for for each theme, by the probability of described user to be identified infection, determines the connection distance between any two users to be identified in described relational network with annexation according to described theme;
    Determining unit, if be also less than preset value for the connection distance having between two users to be identified of annexation, determines that described two users to be identified belong to the user group that described theme is corresponding.
  10. 10. system according to claim 6, is characterized in that, described acquisition module also for, obtain the message content of message to be analyzed;
    Described determination module also for, according to the message content of described message to be analyzed, determine the theme under described message to be analyzed and the probability that belongs to corresponding theme;
    Described acquisition module also for, obtain the theme of probable value maximum in the theme under described message to be analyzed;
    Described determination module also for, by user group corresponding to the theme with described probable value maximum, the user group who is defined as propagating described message to be analyzed, to control described transmission of news to be analyzed.
CN201410194361.4A 2014-05-09 2014-05-09 Social networks Stock discrimination method and system Active CN104035961B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410194361.4A CN104035961B (en) 2014-05-09 2014-05-09 Social networks Stock discrimination method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410194361.4A CN104035961B (en) 2014-05-09 2014-05-09 Social networks Stock discrimination method and system

Publications (2)

Publication Number Publication Date
CN104035961A true CN104035961A (en) 2014-09-10
CN104035961B CN104035961B (en) 2017-06-06

Family

ID=51466731

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410194361.4A Active CN104035961B (en) 2014-05-09 2014-05-09 Social networks Stock discrimination method and system

Country Status (1)

Country Link
CN (1) CN104035961B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111210002A (en) * 2019-12-30 2020-05-29 北京航空航天大学 Multi-layer academic network community discovery method and system based on generation of confrontation network model

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102646134A (en) * 2012-03-29 2012-08-22 百度在线网络技术（北京）有限公司 Method and device for determining message session in message record
CN103399891A (en) * 2013-07-22 2013-11-20 百度在线网络技术（北京）有限公司 Method, device and system for automatic recommendation of network content
CN103684969A (en) * 2012-08-31 2014-03-26 国际商业机器公司 Message handling method and message handling system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102646134A (en) * 2012-03-29 2012-08-22 百度在线网络技术（北京）有限公司 Method and device for determining message session in message record
CN103684969A (en) * 2012-08-31 2014-03-26 国际商业机器公司 Message handling method and message handling system
CN103399891A (en) * 2013-07-22 2013-11-20 百度在线网络技术（北京）有限公司 Method, device and system for automatic recommendation of network content

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111210002A (en) * 2019-12-30 2020-05-29 北京航空航天大学 Multi-layer academic network community discovery method and system based on generation of confrontation network model
CN111210002B (en) * 2019-12-30 2022-01-28 北京航空航天大学 Multi-layer academic network community discovery method and system based on generation of confrontation network model

Also Published As

Publication number Publication date
CN104035961B (en) 2017-06-06

Similar Documents

Publication Publication Date Title
US20160307131A1 (en) Method, apparatus, and system for controlling delivery task in social networking platform
CN106850314B (en) Method and equipment for determining user attribute model and user attribute information
DE102017121712A1 (en) Intelligent answers using a device-internal model
WO2014193399A1 (en) Influence score of a brand
CN104081392A (en) Influence scores for social media profiles
CN110020104A (en) News handles method, apparatus, storage medium and computer equipment
CN104750760A (en) Application software recommending method and device
CN104092567A (en) Method and device for confirming influence sequencing of users
DE102017201173A1 (en) LATENCY REDUCTION IN THE FEEDBACK-BASED DETERMINATION OF A SYSTEM PERFORMANCE
CN103136213A (en) Method and device for providing related words
US9602563B2 (en) Selective content delivery in a real-time streaming environment
CN104281669B (en) The appraisal procedure and device of social network user influence numerical value
CN110737432A (en) script aided design method and device based on root list
CN111353860B (en) Product information pushing method and system
CN112800063A (en) Automatic label passing method and device based on data structure
CN111125332B (en) Method, device, equipment and storage medium for calculating TF-IDF value of word
CN104035961A (en) Method and system for recognizing social internet population
CN106446696B (en) Information processing method and electronic equipment
CN109783175B (en) Application icon management method and device, readable storage medium and terminal equipment
US9634978B2 (en) Systems and methods for improving efficiency of electronic transmission of messages
CN108932228B (en) Live broadcast industry news and partition matching method and device, server and storage medium
CN114861029B (en) Public opinion modeling process analysis method, apparatus, electronic device and storage medium
US20230222890A1 (en) TRIGGERING METHOD AND TRIGGERING APPARATUS of INTERVENTION PROMPT ON THE BASIS OF USER SMOKING BEHAVIOR RECORDS
CN110941952A (en) Method and device for perfecting audit analysis model
CN114297381A (en) Text processing method, device, equipment and storage medium

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant