CN108427715A - A kind of social networks friend recommendation method of fusion degree of belief - Google Patents
A kind of social networks friend recommendation method of fusion degree of belief Download PDFInfo
- Publication number
- CN108427715A CN108427715A CN201810109014.5A CN201810109014A CN108427715A CN 108427715 A CN108427715 A CN 108427715A CN 201810109014 A CN201810109014 A CN 201810109014A CN 108427715 A CN108427715 A CN 108427715A
- Authority
- CN
- China
- Prior art keywords
- user
- similarity
- social
- confidence level
- confidence
- 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
- 230000004927 fusion Effects 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 title claims abstract description 23
- 230000006399 behavior Effects 0.000 claims abstract description 27
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 230000011273 social behavior Effects 0.000 claims abstract description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 claims 3
- 239000000284 extract Substances 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 238000000205 computational method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 206010034719 Personality change Diseases 0.000 description 1
- 235000010627 Phaseolus vulgaris Nutrition 0.000 description 1
- 244000046052 Phaseolus vulgaris Species 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
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
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention proposes a kind of social networks friend recommendation method of fusion degree of belief, is related to user's similarity, confidence factor calculates and fusion.The good friend recognized is biased toward according to the recommendation of social networks topology and ignores potential interested good friend, it is biased toward according to interest recommendation and recommends strange user, it is difficult to obtain users to trust, simultaneously both recommendations all do not account for behavior of the user in social networks, the strong influence accuracy of recommendation results, reliability and comprehensive.The present invention propose it is a kind of consider social networks topology, user interest and Social behaviors recommendation method.First, social similarity is calculated according to common neighbours in user social contact network, Interest Similarity is calculated according to keyword, and carry out linear combination.Consider user social contact topology and Social behaviors, calculates relationship confidence level and behavior confidence level, be fused into confidence factor.Finally, similarity and confidence factor are merged, improves similarity calculation confidence level, generate Top N recommendation lists.
Description
Technical field
The present invention relates to data minings and commending system field, good more particularly to a kind of social networks of fusion degree of belief
Friendly recommendation method.
Background technology
With the development of Web2.0, social networks becomes the typical case occurred in Web2.0 mode developments, individual character
Change recommended technology to be also widely used in social networks.Social networks that is to say social network service (Social
Network Service, SNS), it intuitively says and can be described as network-based social activity, development is broadly divided into:E-mail,
BBS, BLOG, Facebook/ Renren Network etc. stage.In social networks, user is by adding good friend, the modes such as concern mechanism
Carry out interaction and Information Sharing, and adding good friend between each other necessarily makes to generate a kind of contact between user, with when
Between development, a kind of good friend's social networks is together formed between user and its good friend and the good friend of good friend, therefrom we can
To recognize the hobby of user, Information Sharing, communicational aspects and its active degree etc..In addition, due to social networks
In user base number it is bigger, and the user in network is the stranger for being short in understanding and trusting between each other mostly, therefore
Friend relation can not possibly be arbitrarily formed, these are all to carry out having to consider the problems of when friend recommendation in social networks.
As the core component of commending system, the research of proposed algorithm is various to push away constantly by the concern of scholars
Algorithm is recommended to emerge one after another, however proposed algorithm has certain limitation, and what is recommended is the object with " physical property " mostly, and
Research in terms of recommendation in relation to social network user or good friend is then relatively fewer.Even if in existing friend recommendation algorithm,
It is mainly based upon social networks topologies and user interest.Recommend to bias toward by social networks topology and recommends to recognize under line
Good friend mainly obtains similarity indices by common friend, and recommends to bias toward by interest and recommend the footpath between fields with same interest
User is given birth to, often recommendation results are difficult to obtain the trust of user.Both recommendations simultaneously all do not account for user in social network
Behavior in network, the accuracy of the result that strong influence is recommended, reliability and comprehensive.
Invention content
In order to overcome above-mentioned defect existing in the prior art, the object of the present invention is to provide a kind of societies of fusion degree of belief
Hand over network friend recommendation method.This method considers social networks topology, user interest and Social behaviors.First, according to
Common neighbours calculate the social similarity of user in the social networks of family, and interest is calculated according to TF-IDF algorithms and cosine similarity
Similarity, and carry out linear combination.Consider user social contact topology and Social behaviors, calculates separately out relationship degree and behavior is set
Reliability, linear combination is at confidence factor.Finally, similarity and confidence factor are merged, improve similarity calculation confidence level,
Generate Top-N recommendation lists.
In order to realize the above-mentioned purpose of the present invention, the present invention provides a kind of social networks friend recommendations of fusion degree of belief
Method includes the following steps:
S1:Calculate user social contact similarity.The bean vermicelli (follower) of user a forms a set, follower
(followee) different sets are formed.Select follower (followee) computational methods, the social phase of user a and user b
It is defined as follows like degree:Wherein, follower (a), follower (b)
Indicate that the concern user set of user a and user b, follower (a) ∩ follower (b) indicate that user a and user b is total respectively
With the set of follower, | follower (a) ∩ follower (b) | indicate the element of the common follower's set of user a and user b
Number.It can similarly obtain, | follower (a) | and | follower (b) | the concern user set of user a and user b is indicated respectively
Element number;
S2:All texts that extraction user issues in social networks, forwards, segment text, calculate word frequency:Wherein, M indicates that total word number of text, m indicate the number that current words occurs in the text;
S3:IDF is calculated against textual value.In order to reduce the adverse effect of " stop-word " to text classification result, need to calculate
Contribution degree of each words for the theme of current text.Some word is more related to text subject, and its weights is higher, and vice versa.
IDF inverse document frequencies provide the calculation of the weights, and formula is:D indicates all texts
Sum, DwIt indicates the number that word w occurs in total in D, works as DwBigger, the weights of w are smaller;
S4:Calculate the keyword weight vector of user.Using the result of calculation of S3 and S4 as input, formula TFIDF is utilized
=TF*IDF obtains the feature vector vector of usera={ (p1,w1),(p2,w2),(p3,w3) ..., wherein piIndicate attribute, wi
Indicate piWeight.
S5:Calculate the Interest Similarity of user.Using the output of S4 as S5 input, obtain user's key term vector it
Afterwards, the similarity between user is calculated using cosine similarity:Wherein
pai, pbiI-th of numerical value of the feature vector of user a and user b is indicated respectively;
S6:Calculated relationship confidence level, the quantity for paying close attention to good friend jointly using user occupy the ratio of family concern good friend's sum
Metric relation confidence level.Formula is as follows:Wherein, follower
(a), follower (b) indicates the concern user set of user a and user b, follower (a) ∩ follower (b) tables respectively
Show that the set of user a and the common followers of user b, follower (a) ∪ follower (b) indicate that user a and user b is paid close attention to
Union;
S7:It is interactive total to occupy family a and user b using mutual-action behavior number between user a and user b for calculating behavior confidence level
Several ratios indicates behavior confidence level.Computational methods are as follows:Wherein,
Action (a, b) indicates mutual-action behavior quantity of the user a to user b, and (b a) indicates mutual-action behavior numbers of the user b to a to action
Amount.Action (a) and action (b) indicate mutual-action behavior total amounts of the user a to user b respectively;
S8:The result of upper S6 and S7 is merged, confidence factor is obtained:Trusab=t η realtiab+ θ n Trust,
Wherein η+θ=1, η=0.1, θ=0.9;
S9:Merge similarity.Social similarity and Interest Similarity are linearly combined.Calculating formula of similarity is such as
Under:Simab=λ socailSimab+βinterestSimab, wherein λ+β=1, λ=0.7, β=0.3.
S10:On the basis of S9, in order to enhance the confidence level of similarity, similarity is modified using certainty factor, together
Shi Tigao recommends satisfaction and novelty degree.Formula is as follows:resultab=Simab+Trustab, wherein SimabIndicate user a and use
The similarity of family b, TrustabIndicate the confidence factor of a and b.According to resultabGenerate Top-N recommendation lists.
This paper presents a kind of social networks friend recommendation methods of fusion degree of belief, and the recommendation of good friend is divided into two steps
Suddenly:User's similarity calculation and user execute the factor and calculate.It is biased toward according to the recommendation of social networks topology to solve
The good friend of understanding and ignore potential friend recommendation interested, biased toward according to interest recommendation and recommend strange user, it is difficult to obtained
The trust of user, while both recommendations all do not account for behavior of the user in social networks, strong influence are recommended
Result accuracy, reliability and comprehensive problem.User's similarity calculation include social similarity, Interest Similarity and
The fusion of the two similarities.Social networks similarity is calculated first with user's common friend, uses for reference TF-IDF algorithm ideas
The weights of user's keyword are calculated, and Interest Similarity is obtained according to cosine similarity, the two similarities are linearly melted
It closes, finds best fusion parameters, λ=0.7 in the present invention, β=0.3.Secondly, it is merged by relationship confidence level and behavior confidence level
The confidence factor of user is obtained, for correcting similarity calculation result.In the present invention, relationship confidence level and behavior confidence level are melted
Conjunction parameter is η=0.1, and fusion similarity finally is added to obtain as a result, carrying out Top-N recommendations by θ=0.9 with confidence factor.
Description of the drawings
The above-mentioned and/or additional aspect and advantage of the present invention, in conjunction with following accompanying drawings to that will become in the description of embodiment
Obviously and it is readily appreciated that, wherein:
Fig. 1 is the overall flow structural schematic diagram of the present invention;
Fig. 2 is the user interest similarity calculation procedure chart of the present invention;
Fig. 3 is user's similarity calculation flow diagram of the present invention;
Fig. 4 is the confidence factor calculation process schematic diagram of the present invention;
Specific implementation mode
The specific implementation of the present invention is further explained in detail below in conjunction with the accompanying drawings.
Fig. 1 is the overall flow structural schematic diagram of the present invention.The present invention includes mainly three parts:User's similarity meter
Calculation, confidence factor calculating and similarity and confidence factor fusion are obtained ranking and are produced by final revised user's similarity
Raw user's recommendation list.First, the social similarity that user is calculated according to common neighbours in user social contact network, according to TF-
IDF algorithms obtain user key words and its vector, calculate Interest Similarity using cosine similarity, and carry out linear combination.It is comprehensive
It closes and considers user social contact topology and Social behaviors, calculate separately out relationship confidence level and behavior confidence level, linear combination is at confidence
The factor.Finally, similarity and confidence factor are merged, improves similarity calculation confidence level, generated Top-N and recommend row.
Fig. 2 is the user interest similarity calculation procedure chart of the present invention.Its main process is as follows:Extraction user is in society first
All texts for handing over publication, forwarding in network, segment text, calculate word frequency:Wherein, M indicates the total of text
Word number, m indicate the number that current words occurs in the text;Then, IDF is calculated against textual value.Its formula is:D indicates the sum of all texts, DwIt indicates the number that word w occurs in total in D, works as DwIt is bigger, the weights of w
It is smaller;Finally calculate the keyword weight vector of user.The feature vector of user is obtained using formula TFIDF=TF*IDF
vectora={ (p1,w1),(p2,w2),(p3,w3) ..., wherein piIndicate attribute, wiIndicate piWeight.
Fig. 3 is user's similarity calculation flow diagram of the present invention.User social contact similarity is calculated first.User a and
The social similarity of user b is defined as follows:Wherein, follower
(a), follower (b) indicates the concern user set of user a and user b, follower (a) ∩ follower (b) tables respectively
Show the set of user a and the common followers of user b, follower (a) ∩ follower (b) | indicate that user a and user b is common
The element number of follower's set.It can similarly obtain, | follower (a) | and | follower (b) | user a and user are indicated respectively
The element number of concern user's set of b;On the basis of calculating Interest Similarity, similarity fusion is finally carried out, is calculated
Formula is as follows:Simab=λ socailSimab+βinterestSimab, wherein λ+β=1, λ=0.7, β=0.3.
Fig. 4 is the confidence factor calculation process schematic diagram of the present invention.Calculated relationship confidence level first is closed jointly using user
The quantity for being poured in friend occupies the ratio-metric relationship confidence level of family concern good friend's sum.Formula is as follows:Then behavior confidence level is calculated, using between user a and user b
The ratio that mutual-action behavior number occupies the interactive sums of family a and user b indicates that behavior confidence level, computational methods are as follows:Then it is merged, obtains confidence factor:Trustab=η
realtionTrustab+θactionTrustab, wherein η+θ=1, η=0.1, θ=0.9;On this basis, for reinforced phase
Like the confidence level of degree, similarity is modified using certainty factor, while improving and recommending satisfaction and novelty degree.Formula is such as
Under:resultab=Simab+Trustab, wherein SimabIndicate the similarity of user a and user b, TrustabIndicate setting for a and b
Believe the factor.According to resultabGenerate Top-N recommendation lists.
Claims (8)
1. a kind of social networks friend recommendation method of fusion degree of belief, which is characterized in that the method includes:User social contact phase
It calculates, user interest similarity calculation and similarity fusion, relationship confidence calculations, behavior confidence calculations and sets like degree
Believe Factor Fusion, finally by the fusion of similarity and confidence factor, generates Top-N recommendation lists.When carrying out user's recommendation,
Consider social networks topology, user interest calculate social similarity and Interest Similarity, two kinds of similarities are melted
Conjunction obtains user's similarity, considers user social contact topology and Social behaviors, calculates separately out relationship degree and behavior confidence level,
Linear combination is modified similarity calculation result, similarity and confidence factor is merged at confidence factor, improves phase
Confidence level is calculated like degree, generates Top-N recommendation lists.Particular content includes:Convert user social contact network to graph structure, one
User is indicated with a node, and the social phase of user is calculated according to the quantity in user social contact network topology, possessing common friend
Like degree;The Text Feature Extraction keyword forwarded in social networks according to user, issued calculates Interest Similarity;Fusion is social similar
Degree and Interest Similarity, are found best fusion parameters by constantly testing, the similarity of user are indicated with this;Utilize user
The quantity of common concern good friend occupies the ratio-metric relationship confidence level of family concern good friend's sum;Using between user mutually emotionally
Condition calculates the behavior confidence level of user;Best confidence factor fusion parameters are found by experiment, two kinds of confidence levels are fused into
For confidence factor;Even if confidence factor is finally utilized to correct similarity as a result, generating Top-N recommendation results.
2. according to the method described in claim 1, it is characterized in that, each user is regarded as a node, if having between user
Contact, then add a line, to obtain the non-directed graph of entire user network between corresponding node.Search for the common pass of user
The good friend of note calculates the social similarity between any two user.
3. according to the method described in claim 1, it is characterized in that, the related content extraction generated in social networks from user
Keyword, these keywords can accurately describe user's feature, preference etc..Using the similarity degree of keyword, user is indicated
Interest Similarity.
4. according to the method described in claim 1, it is characterized in that, connection more based on two common friends of user social contact circle
System is closer, then the thought that the degree of belief of the two users is higher, first extracts paying close attention to jointly for user a and user b
Good friend occupies the ratio-metric relationship confidence level for paying close attention to good friend's sum in family according to the quantity of common concern good friend.
5. according to the method described in claim 1, it is characterized in that, user social contact network behavior includes reminding, forward and commenting on
Three kinds of forms, mutual-action behavior is more between user, and the degree of belief of user is higher.The weight of three kinds of interaction modes is typically different, and is
Simplify and calculate, equal weight is assigned to three kinds of mutual-action behaviors.Family a and use are occupied using mutual-action behavior number between user a and user b
The ratio of the interactive sums of family b indicates behavior confidence level.
6. according to the method described in claim 1, it is characterized in that, by the social similarity of user and Interest Similarity into line
Property fusion indicate user similarity, and by experiment find best fusion parameters.Similarity between user a and user b
It is expressed as:Simab=λ socailSimab+βinterestSimab, wherein λ+β=1, socailSimabIndicate the society between user
Hand over similarity, interestSimabIndicate the Interest Similarity between user.
7. according to the method described in claim 1, it is characterized in that, the relationship confidence level between user a and user b is denoted as
realtionTrustab, behavior confidence level is denoted as actionTrustab, by the relationship confidence level of user, behavior confidence level carries out
Linear fusion indicates the confidence factor of user, and finds best fusion parameters by experiment.Setting between user a and user b
Believe that factor representation is:Trustab=η realtionTrustab+θactionTrustab, wherein+θ=1 η.
8. according to the method described in claim 1, it is characterized in that, in order to enhance the confidence level of similarity, certainty factor is utilized
Similarity is modified, while improving and recommending satisfaction and novelty degree, is denoted as resultab=Simab+Trustab, wherein
SimabIndicate the similarity of user a and user b, TrustabIndicate the confidence factor of a and b.According to resultabTop-N is generated to push away
Recommend list.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810109014.5A CN108427715A (en) | 2018-01-30 | 2018-01-30 | A kind of social networks friend recommendation method of fusion degree of belief |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810109014.5A CN108427715A (en) | 2018-01-30 | 2018-01-30 | A kind of social networks friend recommendation method of fusion degree of belief |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108427715A true CN108427715A (en) | 2018-08-21 |
Family
ID=63156468
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810109014.5A Pending CN108427715A (en) | 2018-01-30 | 2018-01-30 | A kind of social networks friend recommendation method of fusion degree of belief |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108427715A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108876069A (en) * | 2018-09-18 | 2018-11-23 | 哈尔滨工业大学(威海) | A kind of endowment service recommendation method |
CN109146702A (en) * | 2018-08-25 | 2019-01-04 | 彭卫国 | A kind of user's recommended method and system based on guarantee fund |
CN110737837A (en) * | 2019-10-16 | 2020-01-31 | 河海大学 | Scientific research collaborator recommendation method based on multi-dimensional features under research gate platform |
CN110837598A (en) * | 2019-11-11 | 2020-02-25 | 腾讯科技(深圳)有限公司 | Information recommendation method, device, equipment and storage medium |
CN111241423A (en) * | 2020-01-17 | 2020-06-05 | 江西财经大学 | Deep recommendation method and system integrating trust distrust relation and attention mechanism |
CN111984871A (en) * | 2020-09-04 | 2020-11-24 | 广州繁星互娱信息科技有限公司 | Friend recommendation method, friend recommendation display method, device and equipment |
CN112153220A (en) * | 2020-08-26 | 2020-12-29 | 北京邮电大学 | Communication behavior identification method based on social evaluation dynamic update |
CN112559893A (en) * | 2021-01-11 | 2021-03-26 | 引粒网络科技(上海)有限公司 | User recommendation method in internet matching social scene |
CN113051487A (en) * | 2021-04-20 | 2021-06-29 | 安徽师范大学 | Travel group recommendation method based on user trust and social influence |
WO2022205801A1 (en) * | 2021-03-31 | 2022-10-06 | 南方科技大学 | Game friend recommendation method and system based on close contact, server, and storage medium |
CN115357632A (en) * | 2022-07-21 | 2022-11-18 | 广东技术师范大学 | Learner personalized interpretable potential friend recommendation method and device |
CN115865485A (en) * | 2022-11-30 | 2023-03-28 | 上海纽盾科技股份有限公司 | Stranger safety precaution method and system based on meta universe |
CN116796079A (en) * | 2023-06-30 | 2023-09-22 | 深圳市爱彼利科技有限公司 | Data processing method and device for social evaluation |
CN116861101A (en) * | 2023-06-30 | 2023-10-10 | 深圳市爱彼利科技有限公司 | Data processing method and device for social matching |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103279533A (en) * | 2013-05-31 | 2013-09-04 | 北京华悦博智科技有限责任公司 | Method and system for social relationship recommendation |
CN104021233A (en) * | 2014-06-30 | 2014-09-03 | 电子科技大学 | Social network friend recommendation method based on community discovery |
CN104462308A (en) * | 2014-11-27 | 2015-03-25 | 广东小天才科技有限公司 | Social network friend recommending method and system |
CN105468598A (en) * | 2014-08-18 | 2016-04-06 | 大连民族学院 | Friend recommendation method and device |
CN106570090A (en) * | 2016-10-20 | 2017-04-19 | 杭州电子科技大学 | Method for collaborative filtering recommendation based on interest changes and trust relations |
-
2018
- 2018-01-30 CN CN201810109014.5A patent/CN108427715A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103279533A (en) * | 2013-05-31 | 2013-09-04 | 北京华悦博智科技有限责任公司 | Method and system for social relationship recommendation |
CN104021233A (en) * | 2014-06-30 | 2014-09-03 | 电子科技大学 | Social network friend recommendation method based on community discovery |
CN105468598A (en) * | 2014-08-18 | 2016-04-06 | 大连民族学院 | Friend recommendation method and device |
CN104462308A (en) * | 2014-11-27 | 2015-03-25 | 广东小天才科技有限公司 | Social network friend recommending method and system |
CN106570090A (en) * | 2016-10-20 | 2017-04-19 | 杭州电子科技大学 | Method for collaborative filtering recommendation based on interest changes and trust relations |
Non-Patent Citations (2)
Title |
---|
吴昊等: "社交网络中的好友推荐方法研究", 《现代图书情报技术》 * |
王梦佳等: "结合信任和用户关系的微博关注推荐算法", 《计算机应用研究》 * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109146702A (en) * | 2018-08-25 | 2019-01-04 | 彭卫国 | A kind of user's recommended method and system based on guarantee fund |
CN108876069B (en) * | 2018-09-18 | 2021-11-09 | 哈尔滨工业大学(威海) | Endowment service recommendation method |
CN108876069A (en) * | 2018-09-18 | 2018-11-23 | 哈尔滨工业大学(威海) | A kind of endowment service recommendation method |
CN110737837A (en) * | 2019-10-16 | 2020-01-31 | 河海大学 | Scientific research collaborator recommendation method based on multi-dimensional features under research gate platform |
CN110737837B (en) * | 2019-10-16 | 2022-03-08 | 河海大学 | Scientific research collaborator recommendation method based on multi-dimensional features under research gate platform |
CN110837598A (en) * | 2019-11-11 | 2020-02-25 | 腾讯科技(深圳)有限公司 | Information recommendation method, device, equipment and storage medium |
CN111241423B (en) * | 2020-01-17 | 2023-07-25 | 江西财经大学 | Deep recommendation method and system integrating trust-un-trust relationship and attention mechanism |
CN111241423A (en) * | 2020-01-17 | 2020-06-05 | 江西财经大学 | Deep recommendation method and system integrating trust distrust relation and attention mechanism |
CN112153220B (en) * | 2020-08-26 | 2021-08-27 | 北京邮电大学 | Communication behavior identification method based on social evaluation dynamic update |
CN112153220A (en) * | 2020-08-26 | 2020-12-29 | 北京邮电大学 | Communication behavior identification method based on social evaluation dynamic update |
CN111984871A (en) * | 2020-09-04 | 2020-11-24 | 广州繁星互娱信息科技有限公司 | Friend recommendation method, friend recommendation display method, device and equipment |
CN112559893A (en) * | 2021-01-11 | 2021-03-26 | 引粒网络科技(上海)有限公司 | User recommendation method in internet matching social scene |
WO2022205801A1 (en) * | 2021-03-31 | 2022-10-06 | 南方科技大学 | Game friend recommendation method and system based on close contact, server, and storage medium |
CN113051487A (en) * | 2021-04-20 | 2021-06-29 | 安徽师范大学 | Travel group recommendation method based on user trust and social influence |
CN115357632A (en) * | 2022-07-21 | 2022-11-18 | 广东技术师范大学 | Learner personalized interpretable potential friend recommendation method and device |
CN115865485A (en) * | 2022-11-30 | 2023-03-28 | 上海纽盾科技股份有限公司 | Stranger safety precaution method and system based on meta universe |
CN115865485B (en) * | 2022-11-30 | 2024-06-07 | 上海纽盾科技股份有限公司 | Stranger safety precaution method and system based on meta universe |
CN116796079A (en) * | 2023-06-30 | 2023-09-22 | 深圳市爱彼利科技有限公司 | Data processing method and device for social evaluation |
CN116861101A (en) * | 2023-06-30 | 2023-10-10 | 深圳市爱彼利科技有限公司 | Data processing method and device for social matching |
CN116861101B (en) * | 2023-06-30 | 2024-03-29 | 深圳市爱彼利科技有限公司 | Data processing method and device for social matching |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108427715A (en) | A kind of social networks friend recommendation method of fusion degree of belief | |
US8650149B2 (en) | Portable inferred interest and expertise profiles | |
US9171262B2 (en) | Directed expertise level-based discovery system, method, and device | |
US10528574B2 (en) | Topical trust network | |
Korovaiko et al. | Trust prediction from user-item ratings | |
US20130103692A1 (en) | Predicting User Responses | |
Caron et al. | Mixing bandits: A recipe for improved cold-start recommendations in a social network | |
Kommineni et al. | Machine learning based efficient recommendation system for book selection using user based collaborative filtering algorithm | |
Nie et al. | 2-tuple linguistic intuitionistic preference relation and its application in sustainable location planning voting system | |
Son et al. | Trust information network in social Internet of things using trust-aware recommender systems | |
Grzymala‐Busse et al. | Probabilistic rule induction with the LERS data mining system | |
Stan et al. | Recommender systems using social network analysis: challenges and future trends | |
Singh et al. | A survey on the generation of recommender systems | |
Liu et al. | Personalised context-aware re-ranking in recommender system | |
Zhao et al. | AUI&GIV: recommendation with asymmetric user influence and global importance value | |
Najmani et al. | A comparative study on recommender systems approaches | |
Rahim et al. | An efficient recommender system algorithm using trust data | |
Li et al. | A social recommendation method based on trust propagation and singular value decomposition | |
Yang et al. | ✨ Going Beyond Local: Global Graph-Enhanced Personalized News Recommendations | |
Suhaim et al. | Directional user similarity model for personalized recommendation in online social networks | |
Gou et al. | Collaborative filtering recommendation system based on trust-aware and domain experts | |
Pourgholamali | Mining information for the cold-item problem | |
Sun et al. | Personalized dynamic knowledge-aware recommendation with hybrid explanations | |
Gao et al. | Social recommendation with characterized regularization | |
Dixit et al. | Proposed similarity measure using Bhattacharyya coefficient for context aware recommender system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180821 |