CN108470215A - Degree of belief computational methods are obscured in social networking service - Google Patents
Degree of belief computational methods are obscured in social networking service Download PDFInfo
- Publication number
- CN108470215A CN108470215A CN201810196299.0A CN201810196299A CN108470215A CN 108470215 A CN108470215 A CN 108470215A CN 201810196299 A CN201810196299 A CN 201810196299A CN 108470215 A CN108470215 A CN 108470215A
- Authority
- CN
- China
- Prior art keywords
- user
- value
- trust value
- trust
- indicates
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/048—Fuzzy inferencing
-
- 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 belongs to fields of communication technology, and in particular to degree of belief computational methods are obscured in social networking service;Including:Using the familiarity trust value and similarity trust value of user, overall trust value is calculated;The overall trust value is subjected to Fuzzy Calculation according to the membership function of different language grade, obtains maximum membership function value;According to the maximum membership function value, the corresponding language grade of overall trust value is determined;The present invention be directed to actively interaction between user is distinguish, the more careful trust value accurately calculated by interaction less than this phenomenon is passively interacted in social networks;In addition the characteristics of user interest preference being dissolved into trust, realizing the inclined entertainment orientation of social networks now;The output for trusting semantization is realized using membership function;Therefore, the method for the invention can effectively calculate the trust based on user in social networks.
Description
Technical field
The invention belongs to fields of communication technology, and in particular to degree of belief computational methods are obscured in social networking service.
Background technology
Development with computer and network technology and universal, the more and more too busy to get away social networking service of exchange of people.
In actual life, trust is the basis of person to person's exchange.Equally, in social networks, trust is still that social platform is necessary
A part for consideration.That therefore trusts has been calculated as the hot spot of research.
Part research before this is based on social network structure, is mainly used in non-neighbours' circle node users, is saved to non-neighbours
The research of point is based primarily upon two hypothesis:First, trust it is propagated.Second, it is known to trust between neighbor node user.But
It is that, it is difficult to obtain the trust scoring between neighbor node user usually in social networks, user is reluctant to disclosed to good friend
It scores.Exchange and interdynamic between user are the important indexs of social networks, only go to calculate social use from the structure of social networks
The obvious inadequate convincingness of the trust at family, it is inconsistent with the thinking of people in real world " seeing is believing ".
Other part research is based on user behavior, is used for calculating trust by user mutual behavior, but most of research is neglected
Passive exchange depending on user in social networks significantly more than actively exchanges this phenomenon.Part research is using discrete two-value come table
Show trust, i.e., 0 indicates to distrust, 1 indicates to trust, it is clear that method has very big difference with actual life in this, it is difficult to apply.This
Outside, though mostly being indicated using successive value in research before, but intuitive has been lacked.
At present both at home and abroad in terms of the trust there are many research.Canni and Caverlee etc. is proposed using user
Information resources calculate the algorithm of users to trust.Golbeck and Hendler propose TidalTrust, by user to good friend
Scoring, most suitable path is found out with recursive search algorithm to calculate indirect trust.In addition, Golbeck is also further
The problem of having studied user's similitude shows to trust between customer attribute information similitude and user and deposits ability close relationship.On
It states research to be mainly used in non-neighbours' circle node users, two hypothesis is based primarily upon to the research of non-neighbor node:First, letter
The transitivity appointed.Second, it is known to trust between neighbor node user.But it is difficult to obtain neighbours usually in social networks
Trust scoring between node users, user are reluctant to disclosed score to good friend.
Jennifer et al. proposes the trust calculated based on FOAF in social networks between user.But its trust value is
Two-value is trusted, and only 0 and 1;0 indicates to distrust, 1 indicates to trust.Obviously, the representation method of single solution for diverse problems lacks flexibility, this
Discrete two-value, which trusts representation method and actual life, very big difference, it is difficult to apply.
Wang et al. proposes the transaction record based on seller in e-commerce and using two indices k- in social networks
Core and center weights predict the seller of the risk of transaction, but this method must have transaction record deposits
It is being not suitable for and is calculating the trust value between ordinary user.
Yin et al. proposes AUTrust, is used for calculating trust value by the interbehavior of user, interaction is used as social networks
Interbehavior is used for quantifying trusting the tight type for embodying user and social networks by important indicator.But it does not consider
User actively exchanges number much smaller than passive exchange number in social networks.It only individually considers and talks between user and comment etc.
Active exchange of information.
Therefore how under the premise of calculating user and directly trusting this, the contextual informations such as user preference are incorporated and by user
Actively exchange is distinguished with passive exchange, and is used in combination suitable mode to express trust and is asked as the prior art is to be solved
Topic.
Invention content
In view of this, the purpose of the present invention is to provide degree of belief computational methods are obscured in a kind of social networking service, such as
Shown in Fig. 1, including:
Using the familiarity trust value and similarity trust value of user, overall trust value is calculated;
The overall trust value is subjected to Fuzzy Calculation according to the membership function of different language grade, obtains maximum be subordinate to
Functional value;
According to the maximum membership function value, the corresponding language grade of overall trust value is determined.
Further, the calculation formula of the overall trust value is:
T (u, v)=α FT (u, v)+(1- α) ST (u, v)
Wherein, T (u, v) is the overall trust value of user u to user v, and FT (u, v) indicates familiarity trust value;ST(u,v)
Indicate similarity trust value, α is normalized parameter.
Further, the calculation formula of the familiarity trust value is:
FT (u, v)=λ FTa(u,v)+(1-λ)FTp(u,v)
Wherein, FT (u, v) indicates the familiarity trust value of user u to user v, and λ is normalized parameter, FTa(u, v) is indicated
Active familiarity trust value;FTp(u, v) indicates passive familiarity trust value.
Further, the calculation formula of the active familiarity trust value is:
Wherein, A (u → v) indicates the active interaction times of user u to user v;N is the neighbor node set of user u.
Further, the calculation formula of the passive familiarity value of information is:
Wherein,Indicate the passive interaction times of user u to user v;N is the neighbor node set of user u.
Further, the computational methods of the similarity trust value are:
Wherein, ST (u, v) indicates similarity trust values of the user u to user v;aiIndicate the first indicator function, ai=(ui
=vi∩ui,vi≠ 0), that is, work as ui=viAnd ui≠0、vi≠ 0 when being true, ai=1, otherwise ai=0;biIndicate the second instruction letter
Number, bi=(ui≠0∪vi≠ 0), that is, work as ui≠ 0 or vi≠ 0 when being true, bi=1, otherwise, bi=0;uiIndicate u couples of user i-th
The label situation of project, ui=0 indicates unmarked i-th of the project of user u, ui=1 indicates user u i-th of project of label;viTable
Show the label situation of v pairs of i-th of project of user, vi=0 indicates unmarked i-th of the project of user v, vi=1 indicates user's v labels
I-th of project.
Further, the overall trust value is carried out Fuzzy Calculation according to the membership function of different language grade includes:
Wherein, x indicates trust value, flIndicate the membership function of low language grade;fmlLow language grade in expression
Membership function;fmIndicate the membership function of medium language grade;fmhThe membership function of higher language grade in expression;fhIndicate high
The membership function of equal language grade.
Further, the calculation of the maximum membership function value is:
lμν=max (fl,fml,fm,fmh,fh)
Wherein, lμνFor fl,fml,fm,fmh,fhMiddle maximum value;flIndicate the membership function value of low language grade;fmlIt indicates
In low language grade membership function value;fmIndicate the membership function value of medium language grade;fmhHigher language etc. in expression
The membership function value of grade;fhIndicate the membership function value of higher language grade.
Beneficial effects of the present invention:
It is an advantage of the invention that for actively interaction is less than this phenomenon is passively interacted in social networks, by interaction between user
It is distinguish, the more careful trust value accurately calculated.In addition user interest preference is dissolved into trust, realizes society now
The characteristics of handing over network inclined entertainment orientation.Finally, the output for trusting semantization is realized using membership function, more intuitively.Therefore, originally
Invention the method can effectively calculate the trust based on user in social networks.
Description of the drawings
Fig. 1 is fuzzy degree of belief computational methods embodiment flow chart in a kind of social networking service of the present invention;
Fig. 2 is fuzzy degree of belief computational methods preferred embodiment flow chart in a kind of social networking service of the present invention;
Fig. 3 is that root node user A of the present invention actively scheme and passively exchange figure by exchange.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, below in conjunction with attached drawing to of the invention real
The technical solution applied in example is clearly and completely described, it is clear that described embodiment is only that a present invention part is implemented
Example, instead of all the embodiments.
Degree of belief computational methods are obscured in a kind of social networking service of the present invention, as shown in Figure 1, including the following steps:
Using the familiarity trust value and similarity trust value of user, overall trust value is calculated;
The overall trust value is subjected to Fuzzy Calculation according to the membership function of different language grade, obtains maximum be subordinate to
Functional value;
According to the maximum membership function value, the corresponding language grade of overall trust value is determined.
Wherein, the corresponding language grade of overall trust value that is to say that the acquisition modes of Linguistic Value are specifically as shown in Figure 2:
According to user data, the interbehavior of user and the preference information of user are obtained respectively, according to the interaction of user
Behavior finds out familiarity trust value, and the similarity trust value of user is found out according to the preference information of user, to according to above-mentioned two
Kind trust value finds out the overall trust value of user, and the Linguistic Value corresponding to user's overall trust value is obtained according to membership function, from
And an intuitive trust value can be obtained:It specifically includes:
S1:Similarity trusts application:User preference information mainly is used for calculating the trust of similitude generation.User preference
It is similar to illustrate that the hobby between user is similar.User can use the application service of social networks in social networks, thus can produce
Raw a large amount of preference informations, such as the user people paid close attention to recently, the project etc. that scores recently.Table 1 show user to project
The Closing Binary Marker figure of scoring that is to say and project is marked wherein 1 indicates " liking ";0 indicates " not liking ", that is to say pair
Project is not marked, and m indicates project sum;N indicates total number of users.
1 two-value of table mark figure
S2:The calculating that similarity is trusted:Similarity trust value is calculated by correlation adjustment using Jie Kade similarity factors,
Calculation formula is as follows:
Wherein, ST (u, v) indicates similarity trust values of the user u to user v;aiIndicate the first indicator function, ai=(ui
=vi∩ui,vi≠ 0), that is, work as ui=viAnd ui≠0、vi≠ 0 when being true, ai=1, otherwise ai=0;biIndicate the second instruction letter
Number, bi=(ui≠0∪vi≠ 0), that is, work as ui≠ 0 or vi≠ 0 when being true, bi=1, otherwise, bi=0;uiIndicate u couples of user i-th
The label situation of project, ui=0 indicates unmarked i-th of the project of user u, ui=1 indicates user u i-th of project of label;viTable
Show the label situation of v pairs of i-th of project of user, vi=0 indicates unmarked i-th of the project of user v, vi=1 indicates user's v labels
I-th of project.
S3:The application that familiarity is trusted:The trust that familiarity generates is usually that the exchange between user generates.Same real world
Equally, the exchange between user the at most more familiar.However not all user can actively carry out actively with other users
Exchange, user is more the homepage for browsing other good friends, pushes away text, the behaviors such as thumbs up and forward, these are commonly referred to as passive
Exchange.
S4:The classification that familiarity is trusted:In order to distinguish the exchange of the active between user and passively exchange, familiarity is generated
Trust and is divided into the trust for being actively familiar with generating and the passive trust for being familiar with generating.Active familiarity trust value depends on user actively
Number is exchanged, passive familiarity trust value is then passively exchanged number by user and determined.As shown in figure 3, Fig. 3 is that target user A is good
The schematic diagram for the number of friend actively and passively exchanged;It can be seen that user A between other users actively exchange number and by
Dynamic exchange number, such as:It is 4+8 times that user A actively exchanges number with user B;It is 7+8 that user A actively exchanges number with user C
It is secondary;It is 3+7 times that user A actively exchanges number with user D;It is 9+8 times that user A actively exchanges number with user F;User A and use
It is 12+9 times that family B, which passively exchanges number,;It is 3+15 times that user A passively exchanges number with user C.
It is understood that the exchange number of the present invention that is to say interaction times.
S5:The calculating that active familiarity is trusted and passive familiarity is trusted:
Wherein, FTa(u, v) indicates active familiarity trust value;FTp(u, v) indicates passive familiarity trust value;A(u→v)
Indicate the active interaction times of user u to user v;Indicate the passive interaction times of user u to user v;N is user
The neighbor node set of u.
S6:The calculating of familiarity:Active is familiar with trusting by normalized parameter and is passively familiar with belief function formula
It is as follows:
FT (u, v)=λ FTa(u,v)+(1-λ)FTp(u,v)
Wherein, FT (u, v) indicates the familiarity trust value of user u to v, and λ is normalized parameter, FTa(u, v) is indicated actively
Familiarity trust value;FTp(u, v) indicates passive familiarity trust value.
S7:Overall trust value calculates:The present invention according between social network user familiarity and similitude come calculate trust
Degree, the degree of belief formula for obtaining user u to user v are:
T (u, v)=α FT (u, v)+(1- α) ST (u, v)
Wherein T (u, v) is the overall trust value of user u to user v, and FT (u, v) indicates the familiarity of user u to user v
Trust value;ST (u, v) indicates similarity trust values of the user u to user v;α is normalized parameter.
S8:The necessity of fuzzy trust flaw:Trusting has relativity.In order to allow user to be best understood from trust numerical value, adopt
Fuzzy reasoning is carried out with Linguistic Value, numerical value will be trusted and be converted into understandable Linguistic Value.
S9:Linguistic Value grade:It is high, middle it is high, medium, in it is low and low.Language value set is:
L={ l, ml, m, mh, h }
L indicates low language grade;Low language grade during ml is indicated;M indicates medium language grade;Mh is high in indicating
Language grade;H indicates higher language grade
S10:Reach semantization using membership function:Trapezoidal function is chosen as membership function, formula is as follows:
Wherein x indicates that trust value, f () are membership function.
As a kind of realization method:
S11:The algorithm flow that the present invention uses:
S12:Algorithm Analysis:μ is root node user in above-mentioned algorithm, and N is the buddy list of user μ.Main flow is 1-3
Row traversal calculate user μ to good friend's trust value, then on this basis 5-10 rows calculate user the corresponding language of membership function value
Speech value, last 12 row return to maximum membership values.During this during Similarity measures of user, because common to user
The project of scoring is traversed, so the time complexity of the step is O (N2).Familiarity is trusted in calculating, is user's static state
The extraction of data, Algorithms T-cbmplexity are O (N).So the complexity of entire algorithm is O (N2)。
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium may include:ROM, RAM, disk or CD etc..
Embodiment provided above has carried out further detailed description, institute to the object, technical solutions and advantages of the present invention
It should be understood that embodiment provided above is only the preferred embodiment of the present invention, be not intended to limit the invention, it is all
Any modification, equivalent substitution, improvement and etc. made for the present invention, should be included in the present invention within the spirit and principles in the present invention
Protection domain within.
Claims (9)
1. obscuring degree of belief computational methods in a kind of social networking service, which is characterized in that include the following steps:
Using the familiarity trust value and similarity trust value of user, overall trust value is calculated;
The overall trust value is subjected to Fuzzy Calculation according to the membership function of different language grade, obtains maximum membership function
Value;
According to the maximum membership function value, the corresponding language grade of overall trust value is determined.
2. obscuring degree of belief computational methods in social networking service according to claim 1, which is characterized in that the totality
The calculation formula of trust value is:
T (u, v)=α FT (u, v)+(1- α) ST (u, v)
Wherein, T (u, v) is the overall trust value of user u to user v, and FT (u, v) indicates familiarity trust value;ST (u, v) is indicated
Similarity trust value, α are normalized parameter.
3. obscuring degree of belief computational methods in social networking service according to claim 2, which is characterized in that described to be familiar with
Degree trust value calculation formula be:
FT (u, v)=λ FTa(u,v)+(1-λ)FTp(u,v)
Wherein, FT (u, v) indicates the familiarity trust value of user u to user v, and λ is normalized parameter, FTa(u, v) is indicated actively
Familiarity trust value;FTp(u, v) indicates passive familiarity trust value.
4. obscuring degree of belief computational methods in social networking service according to claim 3, which is characterized in that the active
The calculation formula of familiarity trust value is:
Wherein, A (u → v) indicates the active interaction times of user u to user v;N is the neighbor node set of user u.
5. obscuring degree of belief computational methods in social networking service according to claim 3, which is characterized in that described passive
The calculation formula of the familiarity value of information is:
Wherein,Indicate the passive interaction times of user u to user v;N is the neighbor node set of user u.
6. obscuring degree of belief computational methods in social networking service according to claim 2, which is characterized in that described similar
Degree trust value computational methods be:
Wherein, ST (u, v) indicates similarity trust values of the user u to user v;aiIndicate the first indicator function, i.e., as u couples of user
When v pairs of i-th of project mark of i-th of project mark and user, ai=1, otherwise ai=0;biIt indicates the second indicator function, that is, works as
User u and when unmarked v pairs of i-th of project of user, bi=1, otherwise bi=0;M indicates project sum.
7. obscuring degree of belief computational methods in social networking service according to claim 1, which is characterized in that the maximum
The calculation of membership function value be:
fmax=max (fl,fml,fm,fmh,fh)
Wherein, fmaxFor fl,fml,fm,fmh,fhMiddle maximum value;flIndicate the membership function value of low language grade;fmlIt is low in expression
The membership function value of equal language grade;fmIndicate the membership function value of medium language grade;fmhHigher language grade in expression
Membership function value;fhIndicate the membership function value of higher language grade.
8. obscuring degree of belief computational methods in social networking service according to claim 7, which is characterized in that fl,fml,fm,
fmh,fhCalculation formula be respectively:
Wherein, x=T (u, v), T (u, v) indicate the overall trust value of user u to user v.
9. obscuring degree of belief computational methods in social networking service according to claim 7 or 8, which is characterized in that overall
The corresponding language grade of trust value includes:
L=arg max (fmax)
Work as fmax=flWhen, L=l;Work as fmax=fmlWhen, L=ml;Work as fmax=fmWhen, L=m;Work as fmax=fmhWhen, L=mh;When
fmax=fhWhen, L=h;
L indicates low language grade;Low language grade during ml is indicated;M indicates medium language grade;Higher language during mh is indicated
Grade;H indicates higher language grade.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810196299.0A CN108470215B (en) | 2018-03-09 | 2018-03-09 | Fuzzy trust degree calculating method in social network service |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810196299.0A CN108470215B (en) | 2018-03-09 | 2018-03-09 | Fuzzy trust degree calculating method in social network service |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108470215A true CN108470215A (en) | 2018-08-31 |
CN108470215B CN108470215B (en) | 2021-12-24 |
Family
ID=63264304
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810196299.0A Active CN108470215B (en) | 2018-03-09 | 2018-03-09 | Fuzzy trust degree calculating method in social network service |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108470215B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108921413A (en) * | 2018-06-22 | 2018-11-30 | 郑州大学 | A kind of social networks degree of belief calculation method based on user intention |
CN112733170A (en) * | 2021-01-14 | 2021-04-30 | 中南大学 | Active trust evaluation method based on evidence sequence extraction |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101835158A (en) * | 2010-04-12 | 2010-09-15 | 北京航空航天大学 | Sensor network trust evaluation method based on node behaviors and D-S evidence theory |
CN107330461A (en) * | 2017-06-27 | 2017-11-07 | 安徽师范大学 | Collaborative filtering recommending method based on emotion with trust |
-
2018
- 2018-03-09 CN CN201810196299.0A patent/CN108470215B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101835158A (en) * | 2010-04-12 | 2010-09-15 | 北京航空航天大学 | Sensor network trust evaluation method based on node behaviors and D-S evidence theory |
CN107330461A (en) * | 2017-06-27 | 2017-11-07 | 安徽师范大学 | Collaborative filtering recommending method based on emotion with trust |
Non-Patent Citations (4)
Title |
---|
GUANGYU YIN ET AL.: "AUTrust: A Practical Trust Measurement for Adjacent Users in Social Networks", 《2012 SECOND INTERNATIONAL CONFERENCE ON CLOUD AND GREEN COMPUTING》 * |
廖红梅: "无线传感器网络信任机制研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
潘丹: "基于模糊理论的信任模型研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
陈思憬 等: "基于混合好友路径信任度的社交好友推荐算法", 《计算机技术与发展》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108921413A (en) * | 2018-06-22 | 2018-11-30 | 郑州大学 | A kind of social networks degree of belief calculation method based on user intention |
CN108921413B (en) * | 2018-06-22 | 2021-10-26 | 郑州大学 | Social network trust degree calculation method based on user intention |
CN112733170A (en) * | 2021-01-14 | 2021-04-30 | 中南大学 | Active trust evaluation method based on evidence sequence extraction |
CN112733170B (en) * | 2021-01-14 | 2024-01-30 | 中南大学 | Active trust evaluation method based on evidence sequence extraction |
Also Published As
Publication number | Publication date |
---|---|
CN108470215B (en) | 2021-12-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhai et al. | Measures of probabilistic interval-valued intuitionistic hesitant fuzzy sets and the application in reducing excessive medical examinations | |
Himelboim | Social network analysis (social media) | |
Cabrerizo et al. | Fuzzy decision making and consensus: challenges | |
Wang et al. | Choosing response strategies in social media crisis communication: an evolutionary game theory perspective | |
Du et al. | Distributed data privacy preservation in IoT applications | |
Weatherbee | Counterproductive use of technology at work: Information & communications technologies and cyberdeviancy | |
Liu et al. | An extended multiple criteria decision making method based on neutrosophic hesitant fuzzy information | |
CN107294974A (en) | The method and apparatus for recognizing target clique | |
EP3977685A1 (en) | Providing responses in an event-related session | |
CN107194795A (en) | Credit score model training method, credit score computational methods and device | |
Wang et al. | A multidimensional network link prediction algorithm and its application for predicting social relationships | |
CN108470215A (en) | Degree of belief computational methods are obscured in social networking service | |
CN103843281A (en) | Information exchange in the social network environment | |
Khan et al. | Identification of opinion leaders in social network | |
McGeachy | US-China technology competition: Impacting a rules-based order | |
CN112836137B (en) | Character network support degree computing system and method, terminal, equipment and storage medium | |
Shukla et al. | The great resignation: an empirical study on employee mass resignation and its associated factors | |
Aleinikov et al. | The information efficiency of new knowledge production in the era of network communications | |
Nandi et al. | An efficient link prediction technique in social networks based on node neighborhoods | |
Perez et al. | A dynamic approach to detecting suspicious profiles on social platforms | |
CN113297500B (en) | Social network isolated node link prediction method | |
CN115712772A (en) | Topic propagation prediction method based on topic association | |
CN111078820B (en) | Side weight prediction method based on weight symbol social network embedding | |
Karwowski et al. | Evidence of Chaos in Human Emotions Expressed in Tweets. | |
Chipidza | Who is our Paul Erdös? An analysis of the information systems collaboration network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |