CN101770487A - Method and system for calculating user influence in social network - Google Patents
Method and system for calculating user influence in social network Download PDFInfo
- Publication number
- CN101770487A CN101770487A CN200810306546A CN200810306546A CN101770487A CN 101770487 A CN101770487 A CN 101770487A CN 200810306546 A CN200810306546 A CN 200810306546A CN 200810306546 A CN200810306546 A CN 200810306546A CN 101770487 A CN101770487 A CN 101770487A
- Authority
- CN
- China
- Prior art keywords
- user
- influence power
- behavior
- influence
- module
- 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.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention relates to the field of a social network and is applied to a network system with social elements. The invention discloses a method and a system for calculating user influence in a social network, and in the method, the behaviors of users are analyzed so that a model of the behavior of influence spread among users is established and the mold is used for identifying the most influential people in the social network. The system records user access logs in real time. The method comprises the following steps that: (1) an influence behavior extracting module filters the interactive behaviors in the logs; (2) an influence spread calculating module determines the influence of the current behavior spread according to the type and the actor of the behavior and the emotion tendentiousness analysis of the relative content; (3) an influence updating module mergers the data generated through the increment; (4) a user ranking module incrementally calculates the most influential people; and (5) a service module appoints a target user range and sequences the target users according to the influence. The method and the system can dynamically reflect the change condition of the user influence according to the interactive behaviors of the users in the social network.
Description
Technical field
The present invention relates to the computing method of influence power in a kind of social networks, more particularly, is a kind of influence power computing method based on user behavior analysis, and based on the personalized user ranking system of influence power.
Background technology
Social networks is a kind of internet, applications service, be intended to help people to set up social network, mainly provide and write blog, passed photo, design personal space, function such as listen to the music, play games, the daily behavior of virtual people in reality in network environment satisfies the user and shows the oneself quickly and easily, links up the demand of sharing.Compare with legacy network, the characteristics of social networks are that the user participates in, and promptly encourage the common create contents of numerous netizens.Social networks has been expanded the definition of content, except traditional written form, has also increased a kind of content type of mutual-action behavior newly.
Influence power has been described other people ability of a personal influence, can measure with the attention rate that this people is subjected to, and is subjected to attention rate high more, and its influence power is just big more.Wherein " people " is the common name of unit of account, can be a nature person, also can be a colony with same alike result, for example photography hobby group, the student of Tsing-Hua University etc.Shown in Figure 1 is the synoptic diagram of influence power in the social networks, uses head portrait thickness big or small and arrow to represent the granularity of user force and propagation thereof among the figure.
Influence power has positive and negative branch.For example Liu Dehua among the star in amusement circle and Miss lotus, two people are the focuses of medium and public attention, but are paid close attention to different in kind.Clearly, from the angle of commerce, the influence power of Liu Dehua is more valuable.
The application of influence power is ubiquitous, mainly shows as in internet arena:
A kind of means that the internet becomes influences public opinion that control public opinion have been undisputable fact.The people that influence power is high serves as leader of opinion's role usually in public opinion is propagated, controlled these people, has also just controlled network public opinion effectively.
The ad sales influence power is the major criterion of advertisement position value assessment in the advertising platform.For advertiser, the influence power of media platform is high more, and the advertisement of its issue is paid close attention to more.
User investigation is before new product release, and the common hope of producer invites the user freely to experience among a small circle, and the investigation user is to the suggestion of new product.If experiencing the user is the higher people of influence power among the target customer, experiencing the result so can more have reference value.Simultaneously, compare with the official advertisement, these people's experience report is also more convincing.
The information recommendation influence power is to judge that a people is subjected to effective standard of the degree of recognition.Operation department can this people who recommends some to be approved to the new registration user for foundation.In conjunction with attributes such as occupation, sex, ages, can be from a plurality of dimensions to user's rank, for example the most influential student, white collar, singer or composer, women etc. can recommend corresponding user group respectively.
The calculating of influence power is mainly based on following two kinds of methods at present:
The user estimates this method and calculates other people usually to this evaluation of user number of times, for example certain integrated value (for example linear sum) of space message number, blog message number, photo message number.This method is equal to the author who treats each message, does not also distinguish the emotion tendency of evaluation content.Because the user can also remove the relevant message of oneself at any time, so the influence power accuracy rate that this method is calculated is not high.
The user can add as a friend own cup of tea in the human connection network social intercourse network, even adds as the close friend, thereby forms a human connection relational network.Utilize the level of coverage of everyone network of personal connections to calculate its influence power based on the influence power computing method of human connection network, do not consider the characteristics of social networks, promptly the user may not be ready to indiscriminately imitate fully the relationship among persons in the reality.In stranger's social networks, consider that for privacy the content that exchanges between the acquaintance is limited, in some aspects, people are willing to that more the purpose stranger opens your heart.
In sum, all there is shortcoming separately in above dual mode, or does not distinguish the difference that different user is propagated influence power, or does not consider the negative sense influence power, or the hypothesis user indiscriminately imitates reality on the net fully.Existing influence power computing method design is fairly simple, has ignored cheating factor wherein, usually makes user's ranking result unpredictable deviation occur, causes the user that the satisfaction of website is descended.
Summary of the invention
Purpose of the present invention promptly is to propose influence power New Calculation Method in a kind of social networks at the shortcoming that exists in the above-mentioned prior art, can reflect in real time accurately that the influence power in the social networks changes.
In order to achieve the above object, the invention provides a kind of influence power computing system, comprise influence power behavior extraction module, influence power model computation module and three modules of user's ranking module based on user behavior.Wherein:
Influence power behavior extraction module, the analysis user access log extracts mutual-action behavior wherein, comprise action and two kinds of forms of literal, some do not relate to third-party daily record to remove oneself's visit and editor's personal information etc., get rid of some special numbers of the account, as webmaster, the interior number of the account etc. of promoting of standing.
The influence power model computation module, at first the influence power propagation data that produces according to increment upgrades existing influence power model.And then standardization user's influence power, making the influence power value is a floating number between 0 to 10, thereby makes each result of calculation have comparability.
User's ranking module, according to up-to-date influence power model, the people that the incremental computations influence power is the highest and pay close attention to others people most.
Further, described system also comprises:
The user behavior collection module is used for the operation behavior of recording user access websites.Usually way is to implant one section JS code on the page, carries out this code when user to access pages, the behavior of recording user.Described behavior comprises two aspects: action and literal.As sending out mail action and Mail Contents, browsing blog action and Blog content etc.
The influence power spread calculating module, according to the actor of action and the type of action, and identification of the name in the related text content and emotional orientation analysis, obtain the influence power that this behavior is propagated.In addition, this module also has anti-cheating function, and main rule is as follows:
Propagate the influence power summation that user of the upper limit propagates in a computation period, can not surpass the influence power of self.
Someone repeatedly visits same user the decay rule in a computation period, the influence power of its transmission decays successively.If equal set threshold number, then reduce to 0; If surpass set threshold number, then can produce the effect of propagating negatively influencing power.
Effectively the number of the account hour of log-on number of the account that surpasses 1 computation period could participate in influence power calculating, the influence power initial value of each number of the account and its personal information to enrich degree relevant.
The deletion number of the account is if certain number of the account is deleted, when deleted, and the influence power cancel all of its propagation or acceptance.
Business module according to service needed, defines the scope of " people " in the influence power notion, obtains concrete crowd's user force rank.
In order to achieve the above object, the present invention also provides a kind of influence power computing method, comprises the steps:
(1) after the user logins the social network sites system, the behavioural information of the real time record user of system in this session, and it is stored on the log server.
Further, described user behavior information comprises information such as user ID, IP, COOKIE, access links, access time.
When (2) user finishes this session or at the fixed time, system can the analysis user access log, removes some frustrating behaviors, keeps those and the relevant behavior of influence power propagation.
Further, described frustrating behavior comprises the operation that oneself's visit, editor's personal information, upload pictures etc. and other people have nothing to do.
(3) in the daily record that step (2) produces, name and content ID with wherein convert user ID correspondingly to, still will discard for the daily record of not finding respective user ID.
Further, the relation of described content ID and user ID is meant affiliated relation, for example blog and blog author, mail and mail sender etc.
(4) type of action of every daily record, actor's self influence power, the factors such as emotion tendency of content of text are analyzed in the daily record of handling at step (3), calculate the influence power that this mutual-action behavior is propagated.
Further, described type of action factor is meant that the influence power that dissimilar behaviors is propagated is also different, and its value is directly proportional with the cost that the behavior of sending need be paid.For example send out the mail ratio and browse space more elapsed time and energy, we just think that sending out mail gives user's first so, and are bigger than the influence power of the spatial transmission of browsing first.In the internet, the page number that can need click with the behavior and whether produce word content and weigh.Using page clicks to be the basic calculation user force simply and effectively, is one of innovative point of the present invention.
Further, described actor's influence power factor is meant that the behavior of same type is made by different people, and its influence degree of propagating also is different.Actor self influence power is big more, and the influence power of propagation is just big more.Its quantization step is as follows:
Suppose 1: in the one-period, maximum n time of people's accessible page;
Suppose 2: in the one-period, the maximum m of two person-to-person interbehaviors (m<=n) inferior;
Suppose 3: everyone influence power PR=s of initialization, s are little positive numbers.This initial value is represented the potentiality that the user is paid close attention to, with subscriber data to enrich degree relevant.
So, the x time same individual of visit of user i, the influence power of being propagated is:
Wherein, PR
iIt is the influence power of user i.
Further, described emotional orientation analysis is meant the emotion tendency according to word in the behavior related text content, the positive negativity that decides this influence power to propagate.The main foundation of emotional orientation analysis is an emotion vocabulary dictionary.
(5) result who produces according to step (4) upgrades the user force model, comprises the anti-cheating processing and the result specification processing in later stage in early stage.
Further, described anti-cheating is handled, and is meant to judge whether the actor is deleted, and whether hour of log-on surpasses 1 computation period, according to the final influence power of propagating of decay rule formula calculating, and upgrade actor's concern user list and be subjected to person's quilt to pay close attention to user list.
Further, described result specificationization was meant in the model modification later stage, and to all users' influence power result treatment, value is certain floating number between 0 to 10, made that the result before and after upgrading has comparability.
(6) last, according to targeted customer's scope of concrete professional appointment, select the wherein the highest top n people of influence power.
Further, described business is meant, market department or operation department be according to the target group of concrete service definition, and for example 5000 yuan of Beijing, women, income are with first-class filtercondition.
Compared with prior art, the present invention has following remarkable advantage:
(1) according to dissimilar behaviors, different actors estimates the influence power of behavior, can describe the situation of change of influence power more accurately.
(2) integrated anti-cheating function can suppress those effectively and use fraudulent meanses to obtain the people of high-impact, makes influence power more accurate.
(3) scope of application of the present invention is extensive, can be applied on all websites with social networks element, for example portal website, blog, forum, SNS community, electronic transaction web station system etc.
Description of drawings
Fig. 1 is the synoptic diagram of influence power of the present invention;
Fig. 2 is the structural drawing of influence power computing system of the present invention;
Fig. 3 is the process flow diagram of influence power computing method of the present invention.
Embodiment
Below in conjunction with accompanying drawing and instantiation the present invention is done further introduction, but not as a limitation of the invention.
Figure 2 shows that the structural drawing of influence power computing system of the present invention.As seen from the figure, this system comprises 6 modules: user access logs collection module 1, influence power behavior extraction module 2, influence power spread calculating module 3, influence power model modification module 4, user's ranking module 5, business module 6, wherein:
User access logs collection module 1 is used to collect the operation behavior of user on social network sites, for example edits personal information, writes blog, uploading pictures, listens to the music etc.Usually way is to implant one section JS code on the page, carries out this code when user to access pages, the behavior of recording user.This module also is divided into operation behavior two classes: literal and action are stored in respectively on content server and the log server.
Influence power behavior extraction module 2 reads log database, analyzes every record wherein.Remove self-visit behavior and the behavior relevant, keep the mutual-action behavior between the user with special number of the account.Official's number of the account that wherein special number of the account is the website or the activity that is used to promote, advertisement number of the account.
Influence power spread calculating module 3 to the word content part in the module 2, is finished name to the mapping of user ID according to biographical dictionary, and analyzes the emotion tendency of literal expression according to the emotion dictionary, obtains the positive negativity that this influence power is propagated.Last according to the type of behavior and actor's influence power, the size that decides this influence power to propagate.
Influence power model modification module 4, the influence power increment delta data according to module 3 produces upgrades the influence power model, mainly comprises the anti-cheating identification and the standardization of the influence power in later stage in early stage here.A user repeatedly visits same user, and the influence power of being propagated descends successively, surpasses predetermined threshold, also will become negatively influencing power.The concrete decay rule formula of seeing in the summary of the invention that calculates.
User's rank computing module 5, to the user according to the influence power descending sort.According to the range of application difference, can be divided into two classes:, calculate the people of the most influential people and enthusiasm at all users; At each user, calculate people who pays close attention to me most and the people that I pay close attention to most.Owing to all can produce the daily record data of magnanimity in the social networks usually every day, so the calculating here must be adopted the incremental computations algorithm.
In the influence power computing system of the present invention, user access logs is collected model 1 needs executed in real time, and other steps then can regularly be carried out, and also can carry out at any time in use.Fig. 3 is the process flow diagram of influence power computing method, and the step when at every turn calculating is as follows:
Step S1, influence power behavior extraction comprises attributes such as user ID, time, access links in the daily record that this step need be handled, be typical several capable daily record data fragment below for example:
Sequence number | User ID | Access time | Access links |
(1) | 1300935775 | 20080901121001 | http://messaging.myspace.cn/?friendid=1300000005 |
(2) | 1300935775 | 20080901121005 | http://blog.myspace.cn/e/403037616.htm |
(3) | 1300935775 | 20080901121011 | http://www.myspace.cn/1300935775 |
(4) | 1300935775 | 20080901121100 | http://www.myspace.cn/1304548487 |
(5) | 1300000001 | 20080901121319 | http://messaging.myspace.cn/?friendid=1305605803 |
In the wherein daily record (5) 1300000001 is official control person's numbers of the account, and daily record (3) is that the oneself browses the space, so through step S1, the last daily record relevant with influence power is (1) (2) (4).
Step S2, influence power is propagated calculation procedure, because daily record (1) (2) relates to word content, thus need to handle the name that occurs in the literal, and it is corresponded on the existing number of the account.Suppose to have talked about user 1300935776 in the mail in the daily record (1), the blog author in the daily record (2) is user 1309111111; The emotional orientation analysis of all records all is a forward.Result after handling so is:
The actor | Be subjected to the person | Weight | Tendentiousness |
1300935775 | 1300000005 | 3 | Just |
1300935775 | 1300935776 | 2 | Just |
1300935775 | 1309111111 | 1 | Just |
1300935775 | 1304548487 | 1 | Just |
Step S3, influence power model modification step if find the x time visit of user's first second, will be multiplied by penalty factor to propagation values so when upgrading, obtain the final influence power propagation values of this behavior:
Result after handling so is:
The actor | Be subjected to the person | Weight | Tendentiousness | Propagation values |
1300935775 | 1300000005 | 3 | Just | 0.735 |
1300935775 | 1300935776 | 2 | Just | 0.643 |
1300935775 | 1309111111 | 1 | Just | 0.311 |
1300935775 | 1304548487 | 1 | Just | 0.268 |
Again influence power is carried out standardized operation at last, here with the floating number of influence power value scaled down to 0 between 10.
Wherein denominator is an influence power maximal value in the present model.
Step S4, user's rank.Because social network sites has the registered user of magnanimity usually, so this step must adopt the mode of incremental computations, promptly only calculate the vicissitudinous user of influence power in this cycle, and last time is just the user of preceding N name.
Step S5, business module, concrete professional meeting intended target user scope, for example the area is that Beijing, sex are that woman, monthly income are the user more than 5000 yuan.After the targeted customer determined, this module got final product according to the influence power ordering these users again.
Need to prove, renewal to the influence power model can be set at regularly renewal, rather than when each conversation end, for example: upgrade the influence power model according to the user access logs over seven days weekly, though reduced the number of times that influence power is upgraded, but help finding cheating factor wherein, the influence power between the user is propagated calculate can be more accurate.
The present invention is applicable in the various network services with social element, for example portal website, blog, forum, SNS community, electronic transaction web station system etc.
Claims (9)
1. influence power computing system based on user behavior analysis is characterized in that this system comprises: influence power behavior extraction module, influence power model computation module, user's ranking module, wherein:
Influence power behavior extraction module, the analysis user access log extracts mutual-action behavior wherein, comprises action and two kinds of forms of literal, removes the daily record that relates to special number of the account;
The influence power model computation module, incremental update influence power model, each user's that standardizes influence power makes the floating number of influence power value between 0 to 10;
User's ranking module is according to up-to-date influence power model, incremental computations user's rank.
2. the system as claimed in claim 1 is characterized in that, described system also comprises:
The user behavior collection module, the operation behavior when being used for the recording user access websites.
The influence power propagation module, according to the actor of action and the type of action, and identification of the name of related text content and emotional orientation analysis, obtain the influence power that this behavior is propagated.
Business module, the scope of " people " in the definition influence power notion, user's rank of calculating target group.
3. the system as claimed in claim 1 is characterized in that, described system also comprises anti-cheating processing, and main rule is as follows:
Propagate the influence power summation that user of the upper limit propagates in a computation period, can not surpass the influence power of self.
Someone repeatedly visits same user the decay rule in a computation period, the influence power of its transmission decays successively.If equal set threshold number, then reduce to 0; If surpass set threshold number, then can produce the effect of propagating negatively influencing power.
Effectively the number of the account hour of log-on number of the account that surpasses 1 computation period just participates in influence power and calculates, the influence power initial value of each number of the account and its personal information to enrich degree relevant.
The deletion number of the account is deleted as if certain number of the account, from the deleted moment, and the influence power cancel all of its propagation or acceptance.
4. influence power computing method comprise the steps:
(1) after the user logins the social network sites system, the behavioural information of the real time record user of system in this session, and it is stored on the log server.
When (2) user finishes this session or at the fixed time, system can the analysis user access log, removes some frustrating behaviors, keeps those and the relevant behavior of influence power propagation.
(3) in the daily record that step (2) produces, name and content ID with wherein convert user ID correspondingly to, still will discard for the daily record of not finding respective user ID.
(4) type of action of every daily record, actor's self influence power, the factors such as emotion tendency of word content are analyzed in the daily record of handling at step (3), calculate the influence power that this mutual-action behavior is propagated.
(5) result who produces according to step (4) upgrades the user force model, comprises the anti-cheating processing and the result specification processing in later stage in early stage.
(6) last, according to the targeted customer of concrete professional appointment, obtain the highest top n people of influence power.
5. method as claimed in claim 4 is characterized in that:
Described user behavior comprises information such as user ID, IP, COOKIE, access links, access time.
Described frustrating behavior comprises the operation that self-visit, editing data, upload pictures etc. and other people have nothing to do.
The relation of described content ID and user ID is meant the affiliated relation of content.
6. method as claimed in claim 4 is characterized in that, described emotional orientation analysis is meant the emotion tendency according to word in the word content relevant with behavior, the positive negativity that decides this influence power to propagate.
7. method as claimed in claim 4, it is characterized in that described type of action factor is meant that the influence power that dissimilar behaviors is propagated is also different, its value is directly proportional with the cost that the behavior of sending need be paid, and the present invention uses page clicks and whether produces word content and weighs.
8. method as claimed in claim 4 is characterized in that, described actor's influence power factor is meant that the behavior of same type is made by different people, and its influence degree of propagating also is different.
9. method as claimed in claim 4 is characterized in that, the quantization step that influence power is propagated is as follows:
Suppose 1: in the one-period, maximum n time of people's accessible page;
Suppose 2: in the one-period, the maximum m of two person-to-person interbehaviors (m<=n) inferior;
Suppose 3: initial everyone influence power PR=s, s is a little positive number.Initial value represents that this user attracts the potentiality of paying close attention to, with subscriber data to enrich degree relevant.
So, the x time same individual of visit of user i, the degree of concern of being propagated is:
Wherein, PRi is the influence power of user i.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810306546A CN101770487A (en) | 2008-12-26 | 2008-12-26 | Method and system for calculating user influence in social network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810306546A CN101770487A (en) | 2008-12-26 | 2008-12-26 | Method and system for calculating user influence in social network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101770487A true CN101770487A (en) | 2010-07-07 |
Family
ID=42503352
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200810306546A Pending CN101770487A (en) | 2008-12-26 | 2008-12-26 | Method and system for calculating user influence in social network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101770487A (en) |
Cited By (40)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102567340A (en) * | 2010-12-09 | 2012-07-11 | 腾讯科技(深圳)有限公司 | Method and device for filtering Microblog information |
CN102663101A (en) * | 2012-04-13 | 2012-09-12 | 北京交通大学 | Sina microblog-based user grade sequencing algorithm |
CN102819603A (en) * | 2012-08-17 | 2012-12-12 | 广州多益网络科技有限公司 | Ranking guide interacting system and method |
CN102890696A (en) * | 2011-02-10 | 2013-01-23 | 微软公司 | Social network based contextual ranking |
CN102945279A (en) * | 2012-11-14 | 2013-02-27 | 清华大学 | Evaluating method and device of influence effect of microblog users |
CN103020116A (en) * | 2012-11-13 | 2013-04-03 | 中国科学院自动化研究所 | Method for automatically screening influential users on social media networks |
CN103064918A (en) * | 2012-12-20 | 2013-04-24 | 无锡清华信息科学与技术国家实验室物联网技术中心 | Specialist seeking method and system designed on basis of social recommendation network and mechanism |
CN103178982A (en) * | 2011-12-23 | 2013-06-26 | 阿里巴巴集团控股有限公司 | Method and device for analyzing log |
CN103200073A (en) * | 2012-01-04 | 2013-07-10 | 微梦创科网络科技(中国)有限公司 | Method and device for information processing |
CN103257982A (en) * | 2012-06-13 | 2013-08-21 | 苏州大学 | Blog search result ranking algorithm based on follow relationship |
WO2013160904A1 (en) * | 2012-04-23 | 2013-10-31 | Hewlett-Packard Development Company L.P. | Influence scores for social media profiles |
CN103455615A (en) * | 2013-09-10 | 2013-12-18 | 中国地质大学(武汉) | Method for sequencing filtering and retrieving WeChat accounts |
CN103490918A (en) * | 2012-06-08 | 2014-01-01 | 北京千橡网景科技发展有限公司 | Method and apparatus for tracking user behavior |
CN103593380A (en) * | 2013-08-06 | 2014-02-19 | 北京爱真心信息科技有限公司 | Attractive user recommending platform of online dating site |
CN103870538A (en) * | 2014-01-28 | 2014-06-18 | 百度在线网络技术(北京)有限公司 | Method, user modeling equipment and system for carrying out personalized recommendation for users |
CN103902690A (en) * | 2014-03-27 | 2014-07-02 | 北京邮电大学 | Method for improving accuracy of influence of user generate content (UGC) information of social network |
CN103984706A (en) * | 2014-04-25 | 2014-08-13 | 北京交通大学 | Computation geometry-based algorithm for acquiring influence between microblog users |
CN104035987A (en) * | 2014-05-30 | 2014-09-10 | 南京邮电大学 | Method for ranking microblog network user influence |
CN104408108A (en) * | 2014-11-18 | 2015-03-11 | 重庆邮电大学 | Hot topic group influence analysis system and method based on grey system theory |
CN104462271A (en) * | 2014-11-25 | 2015-03-25 | 广州华多网络科技有限公司 | Method for mining activity degree of user and relevant equipment |
CN105095419A (en) * | 2015-07-15 | 2015-11-25 | 哈尔滨工程大学 | Method for maximizing influence of information to specific type of weibo users |
CN105630800A (en) * | 2014-10-29 | 2016-06-01 | 杭州师范大学 | Node importance ranking method and system |
CN105848178A (en) * | 2015-01-16 | 2016-08-10 | 中国移动通信集团四川有限公司 | Detection and acquisition method and device of unauthorized resource |
CN106204297A (en) * | 2016-07-11 | 2016-12-07 | 深圳市中北明夷科技有限公司 | A kind of recognition methods closing social propagation opinion leader and device |
CN106780073A (en) * | 2017-01-11 | 2017-05-31 | 中南大学 | A kind of community network maximizing influence start node choosing method for considering user behavior and emotion |
CN107145612A (en) * | 2017-06-23 | 2017-09-08 | 北京邮电大学 | A kind of related influence power user of theme has found and method for tracing |
CN107169873A (en) * | 2017-06-16 | 2017-09-15 | 北京信息科技大学 | A kind of microblog users authority evaluation method of multiple features fusion |
WO2017167115A1 (en) * | 2016-03-31 | 2017-10-05 | 阿里巴巴集团控股有限公司 | User evaluation method, device, and apparatus |
CN107341270A (en) * | 2017-07-28 | 2017-11-10 | 东北大学 | Towards the user feeling influence power analysis method of social platform |
CN107341571A (en) * | 2017-06-27 | 2017-11-10 | 华中科技大学 | A kind of social network user behavior prediction method based on quantization social effectiveness |
CN107451612A (en) * | 2017-07-31 | 2017-12-08 | 陕西识代运筹信息科技股份有限公司 | A kind for the treatment of method and apparatus that targeted customer is determined based on concern relation |
CN107832572A (en) * | 2016-09-14 | 2018-03-23 | 腾讯科技(深圳)有限公司 | User force value generation method and device |
CN104424433B (en) * | 2013-08-22 | 2018-12-11 | 腾讯科技(深圳)有限公司 | A kind of anti-cheat method and related system of application program |
CN109145090A (en) * | 2018-08-31 | 2019-01-04 | 东北大学 | A kind of microblog users emotion influence power analysis method based on deep learning |
CN109190058A (en) * | 2018-10-15 | 2019-01-11 | 北京字节跳动网络技术有限公司 | Method and apparatus for handling information |
CN110020375A (en) * | 2017-12-28 | 2019-07-16 | 沈阳新松机器人自动化股份有限公司 | A kind of appraisal procedure of social network user influence power |
CN110196950A (en) * | 2019-02-28 | 2019-09-03 | 腾讯科技(深圳)有限公司 | Propagate the treating method and apparatus of account number |
CN110738421A (en) * | 2019-10-17 | 2020-01-31 | 西南大学 | method for measuring influence of multi-layer network users based on shortest propagation path |
CN112052995A (en) * | 2020-08-31 | 2020-12-08 | 杭州电子科技大学 | Social network user influence prediction method based on fusion emotional tendency theme |
CN113781250A (en) * | 2020-09-14 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Social media information propagation evaluation method and device |
-
2008
- 2008-12-26 CN CN200810306546A patent/CN101770487A/en active Pending
Cited By (68)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102567340B (en) * | 2010-12-09 | 2016-02-24 | 腾讯科技(深圳)有限公司 | A kind of method and device filtering micro-blog information |
CN102567340A (en) * | 2010-12-09 | 2012-07-11 | 腾讯科技(深圳)有限公司 | Method and device for filtering Microblog information |
CN102890696B (en) * | 2011-02-10 | 2017-05-24 | 微软技术许可有限责任公司 | Social network based contextual ranking |
US9870424B2 (en) | 2011-02-10 | 2018-01-16 | Microsoft Technology Licensing, Llc | Social network based contextual ranking |
CN102890696A (en) * | 2011-02-10 | 2013-01-23 | 微软公司 | Social network based contextual ranking |
CN103178982A (en) * | 2011-12-23 | 2013-06-26 | 阿里巴巴集团控股有限公司 | Method and device for analyzing log |
CN103200073A (en) * | 2012-01-04 | 2013-07-10 | 微梦创科网络科技(中国)有限公司 | Method and device for information processing |
CN103200073B (en) * | 2012-01-04 | 2016-08-10 | 微梦创科网络科技(中国)有限公司 | A kind of information processing method and device |
CN102663101B (en) * | 2012-04-13 | 2015-10-28 | 北京交通大学 | A kind of user gradation sort algorithm based on Sina's microblogging |
CN102663101A (en) * | 2012-04-13 | 2012-09-12 | 北京交通大学 | Sina microblog-based user grade sequencing algorithm |
WO2013160904A1 (en) * | 2012-04-23 | 2013-10-31 | Hewlett-Packard Development Company L.P. | Influence scores for social media profiles |
CN104081392A (en) * | 2012-04-23 | 2014-10-01 | 惠普发展公司,有限责任合伙企业 | Influence scores for social media profiles |
CN103490918B (en) * | 2012-06-08 | 2018-07-24 | 北京千橡网景科技发展有限公司 | Method and apparatus for tracking user behavior |
CN103490918A (en) * | 2012-06-08 | 2014-01-01 | 北京千橡网景科技发展有限公司 | Method and apparatus for tracking user behavior |
CN103257982A (en) * | 2012-06-13 | 2013-08-21 | 苏州大学 | Blog search result ranking algorithm based on follow relationship |
CN102819603A (en) * | 2012-08-17 | 2012-12-12 | 广州多益网络科技有限公司 | Ranking guide interacting system and method |
CN103020116A (en) * | 2012-11-13 | 2013-04-03 | 中国科学院自动化研究所 | Method for automatically screening influential users on social media networks |
CN103020116B (en) * | 2012-11-13 | 2015-12-16 | 中国科学院自动化研究所 | The method of the powerful user of automatic screening on social media network |
CN102945279B (en) * | 2012-11-14 | 2016-02-24 | 清华大学 | The appraisal procedure of microblog users influence power and device |
CN102945279A (en) * | 2012-11-14 | 2013-02-27 | 清华大学 | Evaluating method and device of influence effect of microblog users |
CN103064918A (en) * | 2012-12-20 | 2013-04-24 | 无锡清华信息科学与技术国家实验室物联网技术中心 | Specialist seeking method and system designed on basis of social recommendation network and mechanism |
CN103593380A (en) * | 2013-08-06 | 2014-02-19 | 北京爱真心信息科技有限公司 | Attractive user recommending platform of online dating site |
CN104424433B (en) * | 2013-08-22 | 2018-12-11 | 腾讯科技(深圳)有限公司 | A kind of anti-cheat method and related system of application program |
CN103455615A (en) * | 2013-09-10 | 2013-12-18 | 中国地质大学(武汉) | Method for sequencing filtering and retrieving WeChat accounts |
CN103870538A (en) * | 2014-01-28 | 2014-06-18 | 百度在线网络技术(北京)有限公司 | Method, user modeling equipment and system for carrying out personalized recommendation for users |
CN103870538B (en) * | 2014-01-28 | 2017-02-15 | 百度在线网络技术(北京)有限公司 | Method, user modeling equipment and system for carrying out personalized recommendation for users |
CN103902690A (en) * | 2014-03-27 | 2014-07-02 | 北京邮电大学 | Method for improving accuracy of influence of user generate content (UGC) information of social network |
CN103902690B (en) * | 2014-03-27 | 2017-03-22 | 北京邮电大学 | Method for improving accuracy of influence of user generate content (UGC) information of social network |
CN103984706A (en) * | 2014-04-25 | 2014-08-13 | 北京交通大学 | Computation geometry-based algorithm for acquiring influence between microblog users |
CN104035987A (en) * | 2014-05-30 | 2014-09-10 | 南京邮电大学 | Method for ranking microblog network user influence |
CN104035987B (en) * | 2014-05-30 | 2017-06-30 | 南京邮电大学 | A kind of micro blog network user force arrangement method |
CN105630800B (en) * | 2014-10-29 | 2021-01-15 | 杭州师范大学 | Method and system for ordering node importance |
CN105630800A (en) * | 2014-10-29 | 2016-06-01 | 杭州师范大学 | Node importance ranking method and system |
CN104408108A (en) * | 2014-11-18 | 2015-03-11 | 重庆邮电大学 | Hot topic group influence analysis system and method based on grey system theory |
CN104408108B (en) * | 2014-11-18 | 2018-08-14 | 重庆邮电大学 | Much-talked-about topic group influence power analysis system based on gray system theory and method |
CN104462271B (en) * | 2014-11-25 | 2017-11-03 | 广州华多网络科技有限公司 | The method for digging and relevant device of a kind of user activity |
CN104462271A (en) * | 2014-11-25 | 2015-03-25 | 广州华多网络科技有限公司 | Method for mining activity degree of user and relevant equipment |
CN105848178B (en) * | 2015-01-16 | 2019-12-17 | 中国移动通信集团四川有限公司 | method and device for detecting and acquiring unauthorized resource |
CN105848178A (en) * | 2015-01-16 | 2016-08-10 | 中国移动通信集团四川有限公司 | Detection and acquisition method and device of unauthorized resource |
CN105095419A (en) * | 2015-07-15 | 2015-11-25 | 哈尔滨工程大学 | Method for maximizing influence of information to specific type of weibo users |
CN105095419B (en) * | 2015-07-15 | 2018-08-31 | 哈尔滨工程大学 | A kind of informational influence power maximization approach towards microblogging particular type of user |
WO2017167115A1 (en) * | 2016-03-31 | 2017-10-05 | 阿里巴巴集团控股有限公司 | User evaluation method, device, and apparatus |
CN107292465A (en) * | 2016-03-31 | 2017-10-24 | 阿里巴巴集团控股有限公司 | User's evaluation method, device and equipment |
CN106204297A (en) * | 2016-07-11 | 2016-12-07 | 深圳市中北明夷科技有限公司 | A kind of recognition methods closing social propagation opinion leader and device |
CN107832572A (en) * | 2016-09-14 | 2018-03-23 | 腾讯科技(深圳)有限公司 | User force value generation method and device |
CN106780073A (en) * | 2017-01-11 | 2017-05-31 | 中南大学 | A kind of community network maximizing influence start node choosing method for considering user behavior and emotion |
CN106780073B (en) * | 2017-01-11 | 2021-05-25 | 中南大学 | Social network influence maximization initial node selection method considering user behaviors and emotions |
CN107169873A (en) * | 2017-06-16 | 2017-09-15 | 北京信息科技大学 | A kind of microblog users authority evaluation method of multiple features fusion |
CN107169873B (en) * | 2017-06-16 | 2021-05-11 | 北京信息科技大学 | Multi-feature fusion microblog user authority evaluation method |
CN107145612A (en) * | 2017-06-23 | 2017-09-08 | 北京邮电大学 | A kind of related influence power user of theme has found and method for tracing |
CN107341571B (en) * | 2017-06-27 | 2020-05-19 | 华中科技大学 | Social network user behavior prediction method based on quantitative social influence |
CN107341571A (en) * | 2017-06-27 | 2017-11-10 | 华中科技大学 | A kind of social network user behavior prediction method based on quantization social effectiveness |
CN107341270A (en) * | 2017-07-28 | 2017-11-10 | 东北大学 | Towards the user feeling influence power analysis method of social platform |
CN107341270B (en) * | 2017-07-28 | 2020-07-03 | 东北大学 | Social platform-oriented user emotion influence analysis method |
CN107451612A (en) * | 2017-07-31 | 2017-12-08 | 陕西识代运筹信息科技股份有限公司 | A kind for the treatment of method and apparatus that targeted customer is determined based on concern relation |
CN110020375B (en) * | 2017-12-28 | 2023-06-27 | 沈阳新松机器人自动化股份有限公司 | Evaluation method for influence of social network users |
CN110020375A (en) * | 2017-12-28 | 2019-07-16 | 沈阳新松机器人自动化股份有限公司 | A kind of appraisal procedure of social network user influence power |
CN109145090A (en) * | 2018-08-31 | 2019-01-04 | 东北大学 | A kind of microblog users emotion influence power analysis method based on deep learning |
CN109145090B (en) * | 2018-08-31 | 2021-04-30 | 东北大学 | Microblog user emotion influence analysis method based on deep learning |
WO2020078049A1 (en) * | 2018-10-15 | 2020-04-23 | 北京字节跳动网络技术有限公司 | User information processing method and device, server, and readable medium |
CN109190058A (en) * | 2018-10-15 | 2019-01-11 | 北京字节跳动网络技术有限公司 | Method and apparatus for handling information |
CN110196950A (en) * | 2019-02-28 | 2019-09-03 | 腾讯科技(深圳)有限公司 | Propagate the treating method and apparatus of account number |
CN110196950B (en) * | 2019-02-28 | 2023-10-24 | 腾讯科技(深圳)有限公司 | Processing method and device for propagating account |
CN110738421A (en) * | 2019-10-17 | 2020-01-31 | 西南大学 | method for measuring influence of multi-layer network users based on shortest propagation path |
CN110738421B (en) * | 2019-10-17 | 2023-08-22 | 西南大学 | Multilayer network user influence measuring method based on shortest propagation path |
CN112052995A (en) * | 2020-08-31 | 2020-12-08 | 杭州电子科技大学 | Social network user influence prediction method based on fusion emotional tendency theme |
CN112052995B (en) * | 2020-08-31 | 2023-08-01 | 杭州电子科技大学 | Social network user influence prediction method based on fusion emotion tendency theme |
CN113781250A (en) * | 2020-09-14 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Social media information propagation evaluation method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101770487A (en) | Method and system for calculating user influence in social network | |
US11625753B2 (en) | Implicitly associating metadata using user behavior | |
US9245252B2 (en) | Method and system for determining on-line influence in social media | |
TWI494881B (en) | Adaptive audiences for accessibility and privacy control setting claims in a social networking system | |
US20100257028A1 (en) | Methods and systems for extracting and managing latent social networks for use in commercial activities | |
CN111339404B (en) | Content popularity prediction method and device based on artificial intelligence and computer equipment | |
US20090076887A1 (en) | System And Method Of Collecting Market-Related Data Via A Web-Based Networking Environment | |
Chang et al. | Exploring factors influencing mobile users' intention to adopt multimedia messaging service | |
CN109285075A (en) | A kind of Claims Resolution methods of risk assessment, device and server | |
Zaki et al. | Customer experience analytics: dynamic customer-centric model | |
CN103946886A (en) | Structured objects and actions on a social networking system | |
JP6888557B2 (en) | Communication system and communication control method, and program | |
US20140108156A1 (en) | Methods and systems for extracting and managing latent social networks for use in commercial activities | |
KR102322668B1 (en) | Systme for providing multi-platform service for stimulating creative activity of contents creator | |
JP4967233B2 (en) | Agent-based social simulation device applying profiling means | |
Roos et al. | The effect of links and excerpts on internet news consumption | |
CN109325171A (en) | User interest analysis method and system based on domain knowledge | |
CN117217808A (en) | Intelligent analysis and prediction method for activity invitation capability | |
KR102557299B1 (en) | A method of creating the right marketing strategy for your users | |
Hu et al. | Question recommendation for user-interactive question answering systems | |
KR102674020B1 (en) | Artificial intelligence-based influencer recruitment and recommendation system | |
Vlad et al. | Social media as influence factor of quality | |
Ribeiro | A Decentralized Approach to a Social Media Marketing Campaign: Proof of Concept | |
Vieira et al. | The role of social networks for decision-making about tourism destinations | |
Vodrazka | Impact of content attributes on user engagement in social media |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100707 |