CN105574131A - Social network friend making recommendation method and system based on dynamic community identification - Google Patents
Social network friend making recommendation method and system based on dynamic community identification Download PDFInfo
- Publication number
- CN105574131A CN105574131A CN201510933275.5A CN201510933275A CN105574131A CN 105574131 A CN105574131 A CN 105574131A CN 201510933275 A CN201510933275 A CN 201510933275A CN 105574131 A CN105574131 A CN 105574131A
- Authority
- CN
- China
- Prior art keywords
- corporations
- user
- friend
- relation
- dynamic
- 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
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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The invention discloses a social network friend making recommendation method and system based on dynamic community identification, and belongs to the technical field of the Internet. The system function mainly consists of the following four parts: a community identification module, a community classification module, a friend recommendation module and a result display module. The social network friend making recommendation method based on the dynamic community identification comprises the following steps of: a first step, obtaining a user two-degree friend list from a user design relation database, carrying out community identification, and meanwhile calculating a community correlation index; a second step, classifying obtained communities based on a calculation result of the community correlation index; and a third step, recommending friends according to the community classification result, and displaying a friend recommendation result through the community category. The method is used for finding and recommending possible interested people of the user by the method of dynamic community identification and community attribute analysis.
Description
Technical field
The invention belongs to Internet technical field, be specifically related to a kind of social networks friend-making recommend method of identifying based on dynamic corporations and system.
Background technology
The friend recommendation method of current social networks and software is mainly recommended based on the user's similarity measurement in social networks, representational product and function comprise Tencent QQ can knowable people, Facebook PeopleYouMayKnow etc.The common feature of these methods is: the potential good friend that good friend's number is more is jointly more likely recommended.
But in actual applications, the effect of social networks friend recommendation is poor.Existing social networks friend recommendation method is faced with a predicament: the people of recommendation is familiar with, but is unwilling to add as a friend.There is two problems in this: the potential good friend that (1) common good friend's number is more, the uninterested part of user in ripe corporations of user's participation often.Such as, class of user Jia Liao senior middle school 80% classmate as good friend, so very high with common good friend's number of remaining 20%, but may this 20% remaining user be unwilling to add as a friend; Again such as, the colleague of user Jia Liao unit 80%, but the leader of remaining 20% is unwilling to add as a friend; (2) contrary, the potential good friend that some common good friend's number is less may be because corporations are formed.Such as, the college class of new admission, everybody is in understanding and familiar stage mutually, and so potential good friend common good friend number is not high, is but that user is more interested on the contrary.These reasons cause friend recommendation inaccurate based on similarity measurement jointly.
Summary of the invention
In order to overcome the defect that above-mentioned prior art exists, the object of the present invention is to provide a kind of social networks friend-making recommend method of identifying based on dynamic corporations and system.
The present invention is achieved through the following technical solutions:
The invention discloses a kind of social networks friend-making recommend method identified based on dynamic corporations, comprise the following steps:
Step one, obtains user's two degree of buddy lists, carries out corporations' identification from user's design relation database, calculates corporations' correlation metric simultaneously;
Step 2, based on corporations' correlation metric result of calculation, classifies to obtained corporations;
Step 3, according to corporations' classification results, carries out friend recommendation, and shows friend recommendation result by corporations' classification.
From user's design relation database, obtain user's two degree of buddy lists described in step one, carry out corporations' identification, concrete operations are:
1) obtaining the good friend ID list of targeted customer, is the once buddy list of this user; Obtaining the good friend ID list of each good friend of targeted customer again, is two degree of buddy lists of this user; Merge once buddy list and two degree of buddy lists, repeat if exist, then delete the record in two degree of buddy lists, the list after merging is whole good friend ID list;
2) judge in whole good friend ID list, between any two ID, whether there is friend relation, build adjacency matrix;
3) based on adjacency matrix, corporations' identification is carried out to the network of whole good friend ID list composition, export the corporations that each ID belongs to.
Calculating corporations correlation metric described in step one, refer to the corporations belonged to for each ID calculate corporations' relation density, corporations' relation on average set up duration, user and the Connection Density of corporations and the relation of user and corporations on average sets up duration.
Corporations' correlation metric computing method are as follows:
Friend relation number * 2/ [in corporations ID number * (in corporations ID number-1)] in relation density=corporations of corporations;
The friend relation number set up in duration sum/corporations on average setting up relation in duration=corporations of corporations' relation;
In the Connection Density=user of user and corporations and corporations ID friend relation number/corporations in ID number;
The friend relation that the relation of user and corporations on average sets up ID in duration=user and corporations sets up the friend relation number of ID in duration sum/user and corporations.
It is as follows that step 2 carries out classification concrete operations to corporations:
1) on average set up duration based on corporations' relation density and corporations' relation, corporations be divided into ripe corporations, grow up in corporations and initial corporations;
Wherein, define corporations relations on average set up duration be greater than 180 days be ripe corporations; Corporations' relation on average set up duration be less than or equal to 180 days but be greater than 30 days for grow up in corporations; Corporations' relation on average set up duration be less than or equal to 30 days for initial corporations;
2) on average setting up duration based on user and the Connection Density of corporations and the relation of user and corporations is that corporations classify, and corporations is divided into user to enter corporations, user and is entering corporations and the uncorrelated corporations of user;
Wherein, the Connection Density defining user and corporations be greater than 0.6 entered corporations for user; The Connection Density of user and corporations be greater than 0.1 but be less than or equal to 0.6 entering corporations for user; The Connection Density of user and corporations be less than or equal to 0.1 be the uncorrelated corporations of user.
Carry out friend recommendation according to corporations' classification results described in step 3, concrete operations are:
When corporations are the uncorrelated corporations of user, do not trigger any operation;
When corporations have entered corporations for user, do not trigger any operation;
When corporations are corporations or initial corporations in growing up, and for user entering corporations time, according to incorporator and user common good friend's number number, sort from high to low, recommend front 10 to 100 potential good friends that common good friend's number is the highest;
When corporations are ripe corporations, and for user entering corporations time, according to the common good friend's number with user number, sort from high to low, recommend front 1 to 3 potential good friends that common good friend's number is the highest;
After setting up contact with these potential good friends, then show other potential good friend gradually according to the order of common good friend's number from many to few.
Show friend recommendation result by corporations' classification described in step 3, concrete operations are:
When showing the recommendation results of the corporations formed, user disposablely can see front 10 to 100 potential good friends, and these potential good friends can be added as a friend;
And when showing the recommendation results of corporations entered, user can only see front 1 to 3 potential good friends, other potential good friend's gray scale is shown, does not show complete information, can not operate, the increasing and gradually relax control of the good friend's quantity added in these corporations along with user.
The invention also discloses a kind of social networks friend-making commending system identified based on dynamic corporations, this system comprises:
Corporations' identification module, for obtaining user's two degree of buddy lists from user social contact relational database, carrying out corporations' identification, and calculating corporations' correlation metric;
Corporations' sort module, for classifying to the corporations identified;
Friend recommendation module, for recommending obtained corporations' classification;
Result display module, shows friend recommendation result according to corporations' classification.
Corporations' correlation metric that corporations' sort module calculates based on calculating corporations identification module is classified to the corporations identified.
Friend recommendation module is recommended based on obtained corporations' category result.
Compared with prior art, the present invention has following useful technique effect:
The social networks friend-making recommend method identified based on dynamic corporations disclosed by the invention, first from obtaining user's two degree of buddy lists from user's design relation database, carries out corporations' identification, calculates corporations' correlation metric simultaneously; Secondly, based on corporations' correlation metric result of calculation, obtained corporations are classified; Again, according to corporations' classification results, carry out friend recommendation, for user recommends the member entering corporations and initial corporations, recommendation results reflects the current social interests of user; Again, this method can avoid the old relation of recommending out-of-date corporations to user, avoids recommending it not enter the large magnitude relation of corporations to user, avoids bringing privacy leakage; Finally, friend recommendation result is shown by corporations' classification.The inventive method is identified by dynamic corporations and the method for corporations' attributive analysis, finds and recommends user may interested people; Different from existing friend recommendation algorithm " recommend can knowable people ", the inventive method " recommends the interested people of possibility ", the social interests current for user is recommended, and recommends accuracy rate higher, and avoids excessive user's harassing and wrecking and privacy leakage.
The invention also discloses the system that can realize above-mentioned friend-making recommend method, systemic-function is primarily of following four part compositions: corporations' identification module, corporations' sort module, friend recommendation module and result display module.First, corporations' identification module obtains user's two degree of buddy lists from user social contact relational database, carries out corporations' identification, and carries out the calculating of some corporations' key indexs.Based on result of calculation, corporations' sort module is classified to obtained corporations.Friend recommendation module is recommended based on obtained corporations' classification.Finally, result display module shows friend recommendation result according to corporations' classification.
Accompanying drawing explanation
Fig. 1 is the building-block of logic that the present invention is based on the social networks friend-making commending system that dynamic corporations identify;
Fig. 2 is buddy list key diagram involved in the present invention;
Fig. 3 is the social networks friend-making recommendation results exploded view that the present invention is based on the identification of dynamic corporations.
Wherein, 101 is corporations' identification module; 102 is corporations' sort module; 103 is good friend's recommending module; 104 is result display module.
Embodiment
Below in conjunction with specific embodiment, the present invention is described in further detail, and the explanation of the invention is not limited.
The invention discloses a kind of social networks friend-making commending system identified based on dynamic corporations, as shown in Figure 1, wherein, 101 is corporations' identification module to its functional structure; 102 is corporations' sort module; 103 is good friend's recommending module; 104 is result display module.
Systemic-function is primarily of following four part compositions: corporations' identification module, corporations' sort module, friend recommendation module and result display module.
Based on the social networks friend-making recommend method that dynamic corporations identify, comprise the following steps:
Step one, obtains user's two degree of buddy lists, carries out corporations' identification from user's design relation database, calculates corporations' correlation metric simultaneously;
Step 2, based on corporations' correlation metric result of calculation, classifies to obtained corporations;
Step 3, according to corporations' classification results, carries out friend recommendation, and shows friend recommendation result by corporations' classification.
Concrete example explanation is carried out to each module work below:
1, corporations' identification module
The first step, obtains the good friend ID list of targeted customer, is called the once good friend ID list of this user, as the ABCD in Fig. 2.The good friend ID list of each good friend of user, is called two degree of good friend ID list of this user, the CDEFG in Fig. 2; Merging once good friend ID list and two degree of good friend ID list, as run into duplicate keys, deleting the record in two degree of good friend ID list, as the CD in Fig. 2 is only called as once good friend.List after merging is called whole good friend ID list, as the ABCDEFG in Fig. 2.
Second step, judges whether there is friend relation between any two ID in whole good friend ID list, builds adjacency matrix.Such as, the value of (i, j) position of matrix is 1 represent between i-th ID and a jth ID and there is friend relation; There is not friend relation in 0 representative.
3rd step, based on adjacency matrix, the network of whole good friend ID list composition carry out corporations' identification (CommunityDetection).Corporations identify existing proven technique scheme, specifically can with reference to method ([1] Newman such as common Modularity, M.E.J.2004. " Fastalgorithmfordetectingcommunitystructureinnetworks; " PhysicalReviewE (69:6), p066133.; [2] NewmanMEJ.Modularityandcommunitystructureinnetworks [J] .ProceedingsoftheNationalAcademyofSciences, 2006,103 (23): 8577-8582.).After corporations have identified, export the corporations that each ID belongs to.Such as, result of calculation may be ID1001, and 1002,1003 belong to corporations 1; 1004,1005,1006 belong to corporations 2.
4th step is that each corporations calculate corporations' relation density, corporations' relation on average set up duration, the Connection Density of user and corporations, the relation of user and corporations on average sets up duration.Wherein:
Friend relation number * 2/ (in corporations ID number * (in corporations ID number-1)) in relation density=corporations of corporations
The friend relation number set up in duration sum/corporations on average setting up relation in duration=corporations of corporations' relation
In the Connection Density=user of user and corporations and corporations ID friend relation number/corporations in ID number
The friend relation that the relation of user and corporations on average sets up ID in duration=user and corporations sets up the friend relation number of ID in duration sum/user and corporations.
Be exemplified below, suppose that in corporations, ID comprises 1001,1002,1003,1004 totally 4, the relation of existence comprises 1001-1002 (20 days), 1001-1003 (20 days), 1002-1004 (50 days), 1003-1004 (30 days); There is friend relation (20 days) in user and 1001, there is friend relation (30 days) with 1003.
So can calculate:
Corporations relation density=4*2/ (4*3)=0.67
Corporations' relation on average set up duration=(20+20+50+30)/4=30
Connection Density=the 2/4=0.5 of user and corporations
The relation of user and corporations on average sets up duration=(20+30)/2=25.
2, corporations' sort module
1) first this module is corporations' classification based on the duration of on average setting up of the corporations' relation density calculated, corporations' relation, be divided into ripe corporations, grow up in corporations, initial corporations three class.
A kind of possible sorting technique is:
Corporations' relation is on average set up duration and is greater than 180 days: be called ripe corporations;
Corporations' relation is on average set up duration and is less than or equal to 180 days but is greater than 30 days: be called corporations in growth.
Corporations' relation is on average set up duration and is less than or equal to 30 days: be called initial corporations.
2) this module also on average sets up duration based on the relation of the Connection Density of user and corporations, user and corporations is that corporations classify, and is divided into user to enter corporations, user and is entering corporations and the uncorrelated corporations of user.
A kind of possible method is:
The Connection Density of user and corporations is greater than 0.6: be called that user enters corporations
User is greater than 0.1 with the density that links of corporations but is less than or equal to 0.6: be called that user enters corporations
User is less than or equal to 0.1 with the density that links of corporations: be called the uncorrelated corporations of user.
3, friend recommendation module
When corporations are the uncorrelated corporations of user, do not trigger any operation (secret protection);
When corporations have entered corporations for user, do not trigger any operation (avoiding excessively recommending old relation);
When corporations for corporations in growing up or initial corporations and user is entering corporations time, according to incorporator and user common good friend's number number, sort from high to low, recommend the potential good friend of a greater number that common good friend's number is the highest;
When corporations are ripe corporations and user is entering corporations, according to the number of the common good friend's number with user, sort from high to low, recommend the potential good friend of the lesser amt that common good friend's number is the highest; After setting up contact with these potential good friends, then show other potential good friend gradually according to the order of common good friend's number from many to few.
4, recommendation results display module
The above-mentioned recommendation results calculated is showed user by result display module.See Fig. 3, when showing the recommendation results of the corporations formed, user disposablely can see a large amount of potential good friends, and these potential good friends can be added as a friend; And when showing the recommendation results of the corporations entered, user is only to see the potential good friend of minority, and other potential good friend's gray scale is shown, does not show complete information, also cannot operate, the increasing and gradually relax control of the good friend's quantity added in these corporations along with user.
When showing the recommendation results of the corporations formed, user disposablely can see front 10 to 100 potential good friends, and these potential good friends can be added as a friend;
And when showing the recommendation results of corporations entered, user can only see front 1 to 3 potential good friends, other potential good friend's gray scale is shown, does not show complete information, can not operate, the increasing and gradually relax control of the good friend's quantity added in these corporations along with user.
Claims (10)
1., based on the social networks friend-making recommend method that dynamic corporations identify, it is characterized in that, comprise the following steps:
Step one, obtains user's two degree of buddy lists, carries out corporations' identification from user's design relation database, calculates corporations' correlation metric simultaneously;
Step 2, based on corporations' correlation metric result of calculation, classifies to obtained corporations;
Step 3, according to corporations' classification results, carries out friend recommendation, and shows friend recommendation result by corporations' classification.
2. the social networks friend-making recommend method identified based on dynamic corporations according to claim 1, is characterized in that, from user's design relation database, obtain user's two degree of buddy lists described in step one, carry out corporations' identification, concrete operations are:
1) obtaining the good friend ID list of targeted customer, is the once buddy list of this user; Obtaining the good friend ID list of each good friend of targeted customer again, is two degree of buddy lists of this user; Merge once buddy list and two degree of buddy lists, repeat if exist, then delete the record in two degree of buddy lists, the list after merging is whole good friend ID list;
2) judge in whole good friend ID list, between any two ID, whether there is friend relation, build adjacency matrix;
3) based on adjacency matrix, corporations' identification is carried out to the network of whole good friend ID list composition, export the corporations that each ID belongs to.
3. the social networks friend-making recommend method identified based on dynamic corporations according to claim 2, it is characterized in that, calculating corporations correlation metric described in step one, refer to the corporations belonged to for each ID calculate corporations' relation density, corporations' relation on average set up duration, user and the Connection Density of corporations and the relation of user and corporations on average sets up duration.
4. the social networks friend-making recommend method identified based on dynamic corporations according to claim 3, it is characterized in that, corporations' correlation metric computing method are as follows:
Friend relation number * 2/ [in corporations ID number * (in corporations ID number-1)] in relation density=corporations of corporations;
The friend relation number set up in duration sum/corporations on average setting up relation in duration=corporations of corporations' relation;
In the Connection Density=user of user and corporations and corporations ID friend relation number/corporations in ID number;
The friend relation that the relation of user and corporations on average sets up ID in duration=user and corporations sets up the friend relation number of ID in duration sum/user and corporations.
5. the social networks friend-making recommend method identified based on dynamic corporations according to claim 3, it is characterized in that, it is as follows that step 2 carries out classification concrete operations to corporations:
1) on average set up duration based on corporations' relation density and corporations' relation, corporations be divided into ripe corporations, grow up in corporations and initial corporations;
Wherein, define corporations relations on average set up duration be greater than 180 days be ripe corporations; Corporations' relation on average set up duration be less than or equal to 180 days but be greater than 30 days for grow up in corporations; Corporations' relation on average set up duration be less than or equal to 30 days for initial corporations;
2) on average setting up duration based on user and the Connection Density of corporations and the relation of user and corporations is that corporations classify, and corporations is divided into user to enter corporations, user and is entering corporations and the uncorrelated corporations of user;
Wherein, the Connection Density defining user and corporations be greater than 0.6 entered corporations for user; The Connection Density of user and corporations be greater than 0.1 but be less than or equal to 0.6 entering corporations for user; The Connection Density of user and corporations be less than or equal to 0.1 be the uncorrelated corporations of user.
6. the social networks friend-making recommend method identified based on dynamic corporations according to claim 5, it is characterized in that, carry out friend recommendation according to corporations' classification results described in step 3, concrete operations are:
When corporations are the uncorrelated corporations of user, do not trigger any operation;
When corporations have entered corporations for user, do not trigger any operation;
When corporations are corporations or initial corporations in growing up, and for user entering corporations time, according to incorporator and user common good friend's number number, sort from high to low, recommend front 10 to 100 potential good friends that common good friend's number is the highest;
When corporations are ripe corporations, and for user entering corporations time, according to the common good friend's number with user number, sort from high to low, recommend the potential good friend of first 1 to 3 that common good friend's number is the highest;
After setting up contact with these potential good friends, then show other potential good friend gradually according to the order of common good friend's number from many to few.
7. the social networks friend-making recommend method identified based on dynamic corporations according to claim 6, is characterized in that, show friend recommendation result by corporations' classification described in step 3, concrete operations are:
When showing the recommendation results of the corporations formed, user disposablely can see front 10 to 100 potential good friends, and these potential good friends can be added as a friend;
And when showing the recommendation results of corporations entered, user can only see front 1 to 3 potential good friends, other potential good friend's gray scale is shown, does not show complete information, can not operate, the increasing and gradually relax control of the good friend's quantity added in these corporations along with user.
8., based on the social networks friend-making commending system that dynamic corporations identify, it is characterized in that, this system comprises:
Corporations' identification module, for obtaining user's two degree of buddy lists from user social contact relational database, carrying out corporations' identification, and calculating corporations' correlation metric;
Corporations' sort module, for classifying to the corporations identified;
Friend recommendation module, for recommending obtained corporations' classification;
Result display module, shows friend recommendation result according to corporations' classification.
9. the social networks friend-making commending system identified based on dynamic corporations according to claim 8, is characterized in that, corporations' correlation metric that corporations' sort module calculates based on calculating corporations identification module is classified to the corporations identified.
10. the social networks friend-making commending system identified based on dynamic corporations according to claim 8, it is characterized in that, friend recommendation module is recommended based on obtained corporations' category result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510933275.5A CN105574131B (en) | 2015-12-14 | 2015-12-14 | Social network friend making recommendation method and system based on dynamic community identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510933275.5A CN105574131B (en) | 2015-12-14 | 2015-12-14 | Social network friend making recommendation method and system based on dynamic community identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105574131A true CN105574131A (en) | 2016-05-11 |
| CN105574131B CN105574131B (en) | 2017-04-26 |
Family
ID=55884262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510933275.5A Expired - Fee Related CN105574131B (en) | 2015-12-14 | 2015-12-14 | Social network friend making recommendation method and system based on dynamic community identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105574131B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106201184A (en) * | 2016-06-29 | 2016-12-07 | 腾讯科技（深圳）有限公司 | Edit methods, device and the terminal of a kind of SNS message |
| CN106570082A (en) * | 2016-10-19 | 2017-04-19 | 浙江工业大学 | Friend relationship mining method combining network topology characteristics and user behavior characteristics |
| WO2020052039A1 (en) * | 2018-09-13 | 2020-03-19 | 清华大学 | Optimal social recommendation method and device under limited attention |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090157795A1 (en) * | 2007-12-18 | 2009-06-18 | Concert Technology Corporation | Identifying highly valued recommendations of users in a media recommendation network |
| CN103294733A (en) * | 2012-03-05 | 2013-09-11 | 北京千橡网景科技发展有限公司 | Method and system used for recommending friends to users |
| CN104601438A (en) * | 2014-04-28 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Friend recommendation method and device |
| CN104838410A (en) * | 2015-02-11 | 2015-08-12 | 深圳市仁盟互动网络科技有限公司 | Method and system for establishing friendship based on trust recommendation social networking services |
-
2015
- 2015-12-14 CN CN201510933275.5A patent/CN105574131B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090157795A1 (en) * | 2007-12-18 | 2009-06-18 | Concert Technology Corporation | Identifying highly valued recommendations of users in a media recommendation network |
| CN103294733A (en) * | 2012-03-05 | 2013-09-11 | 北京千橡网景科技发展有限公司 | Method and system used for recommending friends to users |
| CN104601438A (en) * | 2014-04-28 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Friend recommendation method and device |
| CN104838410A (en) * | 2015-02-11 | 2015-08-12 | 深圳市仁盟互动网络科技有限公司 | Method and system for establishing friendship based on trust recommendation social networking services |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106201184A (en) * | 2016-06-29 | 2016-12-07 | 腾讯科技（深圳）有限公司 | Edit methods, device and the terminal of a kind of SNS message |
| CN106570082A (en) * | 2016-10-19 | 2017-04-19 | 浙江工业大学 | Friend relationship mining method combining network topology characteristics and user behavior characteristics |
| CN106570082B (en) * | 2016-10-19 | 2019-11-05 | 浙江工业大学 | A kind of friends method for digging of combination network topology characteristic and user behavior characteristics |
| WO2020052039A1 (en) * | 2018-09-13 | 2020-03-19 | 清华大学 | Optimal social recommendation method and device under limited attention |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105574131B (en) | 2017-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020029590A1 (en) | Sample prediction method and device based on federated training, and storage medium | |
| Ghitza et al. | Deep interactions with MRP: Election turnout and voting patterns among small electoral subgroups | |
| CN103823888B (en) | Node-closeness-based social network site friend recommendation method | |
| CN110992059B (en) | Surrounding string behavior recognition analysis method based on big data | |
| CN105574131A (en) | Social network friend making recommendation method and system based on dynamic community identification | |
| Tan et al. | Optimal detection of influential spreaders in online social networks | |
| He et al. | Identifying user behavior on Twitter based on multi-scale entropy | |
| Wainer et al. | How productivity and impact differ across computer science subareas | |
| CN105589916B (en) | Method for extracting explicit and implicit interest knowledge | |
| CN110825817A (en) | Suspected incidence relation determination method and system for enterprise | |
| CN105447117A (en) | User clustering method and apparatus | |
| WO2016106944A1 (en) | Method for creating virtual human on mapreduce platform | |
| Serov et al. | An experimental approach to reputation in e-participation | |
| CN105574139B (en) | It is a kind of that method and system are recommended based on strange make friends of social networks that double Attraction Degrees are calculated | |
| Lu et al. | Beyond the coverage of information spreading: Analytical and empirical evidence of re-exposure in large-scale online social networks | |
| Jafar et al. | Hybrid fuzzy data clustering algorithm using different distance metrics: a comparative study | |
| Opsahl et al. | Modeling the evolution of continuously-observed networks: Communication in a Facebook-like community | |
| CN112598507A (en) | Excessive credit granting risk prediction system and method based on knowledge graph | |
| Iqbal et al. | Association rule analysis-based identification of influential users in the social media | |
| Zhang et al. | Community detection based on structural balance in signed networks | |
| Ding et al. | A tale of two networks: Characterizing the interplay between online and offline communication networks | |
| Hui et al. | A study on directed neutrosophic social networks | |
| Lee | Detection of political manipulation through unsupervised learning | |
| Sharma et al. | Comparative Analysis of Different Algorithms in Link Prediction on Social Networks | |
| CN113190765B (en) | Opinion leader identification method and device based on social network reconstruction |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170426 Termination date: 20191214 |