CN112069416A - Cross-social network user identity recognition method based on community discovery - Google Patents

Cross-social network user identity recognition method based on community discovery Download PDF

Info

Publication number
CN112069416A
CN112069416A CN202010847650.5A CN202010847650A CN112069416A CN 112069416 A CN112069416 A CN 112069416A CN 202010847650 A CN202010847650 A CN 202010847650A CN 112069416 A CN112069416 A CN 112069416A
Authority
CN
China
Prior art keywords
user
social network
similarity
community
social
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010847650.5A
Other languages
Chinese (zh)
Other versions
CN112069416B (en
Inventor
邢玲
邓凯凯
高建平
吴红海
张明川
吴庆涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Henan University of Science and Technology
Original Assignee
Henan University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Henan University of Science and Technology filed Critical Henan University of Science and Technology
Priority to CN202010847650.5A priority Critical patent/CN112069416B/en
Publication of CN112069416A publication Critical patent/CN112069416A/en
Application granted granted Critical
Publication of CN112069416B publication Critical patent/CN112069416B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9536Search customisation based on social or collaborative filtering
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00Reducing energy consumption in communication networks
    • Y02D30/70Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a cross-social network user identity recognition method based on community discovery, which comprises the steps of firstly crawling respective user data from two social networks, then respectively carrying out community division on the two social networks, calculating the similarity of the communities divided by the two social networks, only calculating the similarity of a user in the community with the highest similarity of the user in the community when calculating the similarity of the user in one social network and the user in the other social network, recording the similarity of the user and the rest users as 0, and matching the users in the two social networks according to the obtained user similarity so as to obtain a user identity recognition result. According to the method, the social network is subjected to community division, and the user identification of the large-scale social network is converted into the user identification of the small-scale social network, so that the user identification process is simplified, and the problem of low user identification rate under the condition of large-scale user data is solved.

Description

Cross-social network user identity recognition method based on community discovery
Technical Field
The invention belongs to the technical field of social networks, and particularly relates to a cross-social-network user identity identification method based on community discovery.
Background
The wide application of the Web 3.0 technology promotes the rapid development of Social Networking (SN), and more users begin to participate in the development and perform information interaction. From the latest statistical report in 2020, there are approximately 24.98 million active users and 11.65 million active users on WeChat on Facebook per month. Due to the differences in the application scenarios and functions of the large social networks, people gradually start to use different social networks to meet their social needs. If people keep communicating with friends nearby through QQ and WeChat; paying attention to hot news through a microblog and a Twitter; establishing a relationship of the figures on the workplace through LinkedIn; by knowing about answering questions and solving questions, the knowledge of things is shared.
User identification is also referred to as user identity resolution. The existing related work basically adopts three types of user information, namely user profile information, network topology information and user behavior information, to identify the user identity. Research based on user profile information focuses mainly on basic information of the user, such as user name, gender, interests, etc. However, with the development of social networks and the improvement of user privacy awareness, the basic information is difficult to obtain and has high cost, and in the identification process, the user information has falsification and has a large influence on the performance of user identity identification. Related research based on network topology information mainly focuses on friend networks of users, and the user identities are identified by using the relationships between the users and neighbor nodes. However, social networks are heterogeneous and some users are reluctant to disclose their friend networks, and therefore, further improvements in identification performance are needed. The research based on the user behavior information mainly focuses on the user release content, and the user information is easy to obtain and has high accessibility compared with other two kinds of information, and what is more important is that the information can be personalized to map the behavior habits of the user. However, the published contents of some users are also sparse, which affects the performance of user identification to some extent.
Although many methods have been proposed in the industry to perform user identification between social networks, research shows that along with the expansion of the user data scale of the social networks, the accuracy of the methods for user identification is reduced to different degrees, and the situation that the accuracy of user identification is negatively related to the user data scale is presented.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a cross-social network user identity recognition method based on community discovery, which simplifies user identification through community division and improves the user identity recognition accuracy rate under the condition of large-scale user data.
In order to achieve the purpose, the cross-social network user identity recognition method based on community discovery comprises the following steps:
s1: when the users in the social network A need to be identified with the same account in the social network B, data of the users are respectively crawled from the social network A and the social network B, and the number of the users in the two social networks is respectively NAAnd NB
S2: respectively carrying out community division on the social network A and the social network B;
s3: calculating the similarity between each community in the social network A and each community in the social network B;
s4: for each user i in the social network a, the similarity with the user in the social network B is calculated by adopting the following method:
firstly, a community a to which a user i belongs in a social network A is obtainediSearching all communities in the social network B and the community aiThe community with the highest similarity is marked as biCalculating to obtain the user i and the community b according to the user data crawled in the step S1iSimilarity of all users in the social network B, and the user i and the community B in the social network BiThe similarity of all other users is marked as 0;
s5: and matching the users in the two social networks according to the similarity between each user in the social network a and each user in the social network B obtained in the step S4, so as to obtain a user identification result.
The invention discloses a cross-social network user identity recognition method based on community discovery, which comprises the steps of firstly crawling respective user data from two social networks, then respectively carrying out community division on the two social networks, calculating the similarity of the communities divided by the two social networks, only calculating the similarity of a user in the community with the highest similarity of the user in the community when calculating the similarity of the user in one social network and the user in the other social network, recording the similarity of the user and the rest users as 0, and matching the users in the two social networks according to the obtained user similarity so as to obtain a user identity recognition result.
According to the invention, the social network is subjected to community division, and the user identification of the large-scale social network is converted into the user identification of the small-scale social network, so that the user identification process is simplified, and the problem of low user identification rate under the condition of large-scale user data is solved.
Drawings
FIG. 1 is a flowchart of an embodiment of a cross-social-network user identification method based on community discovery;
FIG. 2 is an exemplary diagram of a five-type edge;
fig. 3 is an exemplary diagram of a hash table-in and a hash table-out.
Detailed Description
The following description of the embodiments of the present invention is provided in order to better understand the present invention for those skilled in the art with reference to the accompanying drawings. It is to be expressly noted that in the following description, a detailed description of known functions and designs will be omitted when it may obscure the subject matter of the present invention.
Examples
FIG. 1 is a flowchart of an embodiment of a cross-social-network user identification method based on community discovery. As shown in fig. 1, the method for identifying the user identity across social networks based on community discovery of the present invention includes the following specific steps:
s101: acquiring user data:
when the users in the social network A need to be identified with the same account in the social network B, data of the users are respectively crawled from the social network A and the social network B, and the number of the users in the two social networks is respectively NAAnd NB
S102: community division:
the social network A and the social network B are respectively subjected to community division.
For community division of a social network, it is desirable to fully consider network topology and attributes of user nodes themselves. The network topology structure of the user can generally determine the global attribute of the discovered community structure, and the attribute characteristics of the user node play an important role in the process of local fine tuning. The importance of the self-attribute of the node is not fully considered by the existing community discovery algorithm. Although a certain algorithm obtains higher modularity, the problem of wrong user node division can still be caused due to the lack of discussion on the correlation between user nodes. Therefore, in order to solve the above problem, the present embodiment employs a community discovery algorithm based on node similarity to perform accurate community partition on a large-scale user data set. The specific method of the community discovery algorithm based on node similarity in this embodiment is as follows:
for the social network needing community division, the similarity between every two user nodes is calculated respectively, and the calculation formula is as follows:
Figure BDA0002643637180000041
where Sim (i, j) represents the similarity between user node i and user node j in the social network, e1Representing the number of edges directly connected between user node i and user node jNoting that the common neighbor user node set of the user node i and the user node j is phi, e2Representing the number of edges directly connected between user nodes in a set phi of common neighbor user nodes, e3Representing the number of edges in the common neighbor user node set phi, which are directly connected with the user node i and the user node j, e4Representing the number of edges which are directly connected with the user nodes and the user nodes i in the common neighbor user node set phi and the edges which are directly connected with the user nodes and the user nodes j in the common neighbor user node set phi simultaneously, e5Representing the number of edges directly connected between the user node in a common neighbor user node set phi and other user nodes which are not directly connected with the user node i and the user node j, w1、w2、w3、w4、w5Weights preset to indicate the number of edges of different types, and satisfy w1>w2>w3>w4>w5
And carrying out hierarchical clustering on the user nodes according to the calculated similarity of the user nodes, and taking the obtained sub-network formed by the users in each class as a community, thereby finishing the community division of the social network.
FIG. 2 is an exemplary diagram of five types of edges. As shown in FIG. 2, in calculating the similarity between user node i and user node j, e1=1,e2=1,e3=4,e4=1,e 52. By dividing the edges into five types, the relationship among the user nodes and the common neighbor relationship among the user nodes are considered, and the relationship among the multi-hop user nodes is also considered, so that the calculated similarity of the user nodes can be more accurate.
In order to better query a public neighbor node set between nodes, in this embodiment, a hash storage table is used to improve the efficiency of the data storage and query process of the social network, and the specific method is as follows:
and establishing a hash table-in for storing node incident side information and a hash table-out for storing node emergent side information for the social network needing community division. In the hash table-in and hash table-out, the key value (key value) is an object function of the tuple g (i, j) defined by the edge corresponding to the source user node i and the destination user node j. The expression of g (i, j) in this embodiment is:
g(i,j)=j|(i<<16)
where, | represents a bitwise or operator, < < represents a bitwise left shift operator.
And substituting the key code value into a preset hash function to obtain a hash address corresponding to the edge in the hash table. In this embodiment, a Fibonacci hash function is used to avoid occurrence of a large-scale hash collision, which is defined as:
Figure BDA0002643637180000051
wherein x represents a key value, M represents the size of the hash table, and W takes a value of 264-1, phi denotes the golden section ratio.
When the similarity of the user nodes in the social network is calculated, the number of the public user node set and the number of the five types of edges can be obtained by searching the established hash table-in and hash table-out.
Fig. 3 is an exemplary diagram of a hash table-in and a hash table-out. As shown in fig. 3, for the user nodes a and b, the hash table-in stores the information of the incident edge, the hash table-out stores the information of the emergent edge, the hash table stores the information in a hash bucket manner, and the stored information is a triple ((i, j), w)i,j),wi,jRepresenting the weight of the edge between source user node i and destination user node j.
S103: calculating the community similarity:
the similarity of each community in social network a to each community in social network B is calculated.
By dividing social networks including large-scale users, network topologies formed by the same user in different social networks have certain similarity. Therefore, the similarity of the communities can be calculated through the matching degree between the nodes,the optimal community is effectively selected for matching, and therefore user identity recognition is better achieved. The method for calculating the community similarity in the embodiment comprises the following steps: obtaining a plurality of account pairs belonging to the same user from two social networks in advance as seed account pairs, and then calculating the similarity H between the p-th community of the social network A and the q-th community of the social network B according to the following formulapq
Figure BDA0002643637180000052
Wherein, FApRepresents a set of user nodes belonging to a seed account pair in the pth community of social network a, p ═ 1,2, …, MA,MARepresenting the number of communities into which social network A is divided, FBqRepresenting a set of user nodes belonging to a seed account pair in the qth community of social network B, q being 1,2, …, MB,MBThe number of communities obtained by dividing the social network B is represented, and | l represents the number of user nodes in the set.
S104: calculating the similarity of the users:
for each user i in the social network a, the similarity with the user in the social network B is calculated by adopting the following method:
firstly, a community a to which a user i belongs in a social network A is obtainediSearching all communities in the social network B and the community aiThe community with the highest similarity is marked as biAccording to the user data crawled in the step S101, calculating to obtain the user i and the community biSimilarity of all users in the social network B, and the user i and the community B in the social network BiThe similarity of all other users is noted as 0.
In order to better calculate the similarity between user nodes, in this embodiment, three similarities are selected for fusion, which are the similarity of user profile information, the similarity of user network topology information, and the similarity of user behavior information, that is, for user i in social network a and community B in social network BiThe user k calculates the similarity of the user profile information, the similarity of the user network topology structure information and the user k respectivelyAnd (4) behavior information similarity, and then weighting and summing the three similarities by adopting preset weights to obtain the user similarity. The following describes three methods for calculating similarity, respectively:
similarity of user Profile information
The user profile information refers to basic information to be filled in advance when a user registers a social account, the user information is easy to obtain and contains more user attributes, but the influence of the contained attributes on the identification performance in the user identity identification process is large, for example, the display name can accurately map the behavior habit of the user, and the interest cannot well distinguish different accounts. In addition, there is a forgery of part of the user profile information. Therefore, the user information attribute items need to be reasonably selected in the user identification process. The method for calculating the similarity of the user profile information in the embodiment comprises the following steps:
connecting user i in social network A with social network B community BiThe file information of the user k adopts string respectivelyi、stringkIndicating, noting stringiThrough a series of editing steps to stringkThe number of times of transformation of (1) is d (string)i,stringk) Then, the user profile information similarity S of user i and user k1The calculation formula of (i, k) is as follows:
Figure BDA0002643637180000061
where L () denotes the length of the character string.
Similarity of user network topology information
The user network topological structure information refers to a friend network formed by the user in the social network, and the user information is easy to obtain and can be truly mapped to form a friend relationship of the user node on the social network. By analyzing the number of common neighbor nodes shared by different user nodes, the similarity between the user nodes is calculated, and the matching of the user nodes can be better realized. In addition, the social network is subjected to community division through community discovery, the network map structure is further refined, the friend relationship of the user node is clearer, and the user identity identification performance is further improved. The method for calculating the similarity of the user network topological structure information in the embodiment comprises the following steps:
obtaining a plurality of account pairs belonging to the same user from two social networks in advance as seed account pairs, and respectively obtaining a user i in the social network A and a community B in the social network BiNeighbor node set of user k belonging to seed account pairikThen, the similarity S of the user network topology structure information of user i and user k2The calculation formula of (i, k) is as follows:
S2(i,k)=|ik|
similarity of user behavior information.
The user behavior information refers to the sum of behavior information of the user in the social network, and the personalized features of the information are obvious compared with those of the first two types of user information. For example, the content posted by the user on the social network can be clearly mapped to the behavior habit of the user and the geographical position formed by the user on different social networks, and the attributes can be mapped to the real state of the user. The method for calculating the similarity of the user behavior information in the embodiment comprises the following steps:
respectively acquiring user i in social network A and community B in social network BiExtracting a plurality of characteristic parameters, such as the occurrence frequency of a plurality of characteristic words, in the historical release content of the user k to form a behavior vector eta of the user i and the user ki、ηkThen two behavior vectors ηi、ηkThe cosine similarity of the two users i and k is used as the similarity S of the user behavior information of the two users i and k3(i,k)。
S105: matching users:
and matching the users in the two social networks according to the similarity between each user in the social network a and each user in the social network B obtained in the step S4, so as to obtain a user identification result.
In this embodiment, a two-way stable marital matching algorithm is used for user matching, and the specific steps include:
1) and carrying out primary matching on the user accounts according to the user similarity of all the user accounts contained in the social network A and all the user accounts in the social network B to obtain candidate matching pairs.
2) And according to the obtained final grading sequence, carrying out relevant matching on the corresponding user account on the social network A and the first named user account on the social network B. If the corresponding user account on the social network B is not matched with other user accounts on the social network A, matching the user account with the current user account on the social network A; if the user account is matched with other user accounts on the social network A, performing similarity score comparison on all the user accounts matched with the user account, and finally selecting the user account with the highest score to perform final matching;
3) and if the account numbers which are not matched exist in the user data set, returning to the second step until all the user account numbers are matched.
Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, and various changes may be made apparent to those skilled in the art as long as they are within the spirit and scope of the present invention as defined and defined by the appended claims, and all matters of the invention which utilize the inventive concepts are protected.

Claims (6)

1. A cross-social network user identity recognition method based on community discovery is characterized by comprising the following steps:
s1: when the users in the social network A need to be identified with the same account in the social network B, data of the users are respectively crawled from the social network A and the social network B, and the number of the users in the two social networks is respectively NAAnd NB
S2: respectively carrying out community division on the social network A and the social network B;
s3: calculating the similarity between each community in the social network A and each community in the social network B;
s4: for each user i in the social network a, the similarity with the user in the social network B is calculated by adopting the following method:
firstly, a community a to which a user i belongs in a social network A is obtainediSearching all communities in the social network B and the community aiThe community with the highest similarity is marked as biCalculating to obtain the user i and the community b according to the user data crawled in the step S1iSimilarity of all users in the social network B, and the user i and the community B in the social network BiThe similarity of all other users is marked as 0;
s5: and matching the users in the two social networks according to the similarity between each user in the social network a and each user in the social network B obtained in the step S4, so as to obtain a user identification result.
2. The method for identifying the user identity across the social networks according to claim 1, wherein the specific method for community division in the step S2 is as follows:
for the social network needing community division, the similarity between every two user nodes is calculated respectively, and the calculation formula is as follows:
Figure FDA0002643637170000011
where Sim (i, j) represents the similarity between user node i and user node j in the social network, e1Representing the number of edges directly connected between the user node i and the user node j, and recording the common neighbor user node set of the user node i and the user node j as phi, e2Representing the number of edges directly connected between user nodes in a set phi of common neighbor user nodes, e3Representing the number of edges in the common neighbor user node set phi, which are directly connected with the user node i and the user node j, e4Representing simultaneous and commonThe number of the edges which are directly connected with the user nodes and the user nodes i in the neighbor user node set phi and the edges which are directly connected with the user nodes and the user nodes j in the common neighbor user node set phi simultaneously, e5Representing the number of edges directly connected between the user node in a common neighbor user node set phi and other user nodes which are not directly connected with the user node i and the user node j, w1、w2、w3、w4、w5Weights preset to indicate the number of edges of different types, and satisfy w1>w2>w3>w4>w5
And carrying out hierarchical clustering on the user nodes according to the calculated similarity of the user nodes, and taking the obtained sub-network formed by the users in each class as a community, thereby finishing the community division of the social network.
3. The method for identifying the user identity across the social networks according to claim 2, wherein a hash table-in for storing the node incident side information and a hash table-out for storing the node exit side information are established for the social networks which need to be community divided; in the hash table-in and hash table-out, the key value (key value) is an object function of a tuple g (i, j) defined by an edge corresponding to a source user node i and a destination user node j; substituting the key code value into a preset hash function to obtain a hash address corresponding to the edge in a hash table;
when the similarity of the user nodes in the social network is calculated, the number of the public user node set and the number of the five types of edges can be obtained by searching the established hash table-in and hash table-out.
4. The method for identifying identity across social networks according to claim 1, wherein the community similarity in step S3 is calculated by: obtaining a plurality of account pairs belonging to the same user from two social networks in advance as seed account pairs, and then calculating the phase of the p-th community of the social network A and the q-th community of the social network B according to the following formulaSimilarity Hpq
Figure FDA0002643637170000021
Wherein, FApRepresents a set of user nodes belonging to a seed account pair in the pth community of social network a, p ═ 1,2, …, MA,MARepresenting the number of communities into which social network A is divided, FBqA set of user nodes belonging to a seed account pair in the qth community of social network B, q being 1,2, …, MB,MBThe number of communities obtained by dividing the social network B is represented, and | l represents the number of user nodes in the set.
5. The method for identifying users across social networks according to claim 1, wherein the similarity between users in step S4 is calculated as follows:
for user i in social network A and social network B community BiThe user k firstly respectively calculates the similarity of user profile information, the similarity of user network topology information and the similarity of user behavior information, and then carries out weighted summation on the three similarities by adopting preset weights to obtain the user similarity, wherein the calculation method of the user profile information similarity is as follows:
connecting user i in social network A with social network B community BiThe file information of the user k adopts string respectivelyi、stringkIndicating, noting stringiThrough a series of editing steps to stringkThe number of times of transformation of (1) is d (string)i,stringk) Then, the user profile information similarity S of user i and user k1The calculation formula of (i, k) is as follows:
Figure FDA0002643637170000031
wherein, L () represents the length of the character string;
user network topologyThe method for calculating the information similarity comprises the following steps: obtaining a plurality of account pairs belonging to the same user from two social networks in advance as seed account pairs, and respectively obtaining a user i in the social network A and a community B in the social network BiNeighbor node set of user k belonging to seed account pairikThen, the similarity S of the user network topology structure information of user i and user k2The calculation formula of (i, k) is as follows:
S2(i,k)=|ik|
the method for calculating the similarity of the user behavior information comprises the following steps: respectively acquiring user i in social network A and community B in social network BiExtracting a plurality of characteristic parameters in the historical release content of the user k to form a behavior vector eta of the user i and the user ki、ηkThen two behavior vectors ηi、ηkThe cosine similarity of the two users i and k is used as the similarity S of the user behavior information of the two users i and k3(i,k)。
6. The method for identifying a user across social networks according to claim 1, wherein the step S5 is implemented by using a two-way stable marital matching algorithm for user matching.
CN202010847650.5A 2020-08-21 2020-08-21 Cross-social network user identity recognition method based on community discovery Active CN112069416B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010847650.5A CN112069416B (en) 2020-08-21 2020-08-21 Cross-social network user identity recognition method based on community discovery

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010847650.5A CN112069416B (en) 2020-08-21 2020-08-21 Cross-social network user identity recognition method based on community discovery

Publications (2)

Publication Number Publication Date
CN112069416A true CN112069416A (en) 2020-12-11
CN112069416B CN112069416B (en) 2022-09-02

Family

ID=73658810

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010847650.5A Active CN112069416B (en) 2020-08-21 2020-08-21 Cross-social network user identity recognition method based on community discovery

Country Status (1)

Country Link
CN (1) CN112069416B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112800345A (en) * 2021-02-03 2021-05-14 安徽大学 Community role-aware user demand active prediction method and system
CN114168733A (en) * 2021-12-06 2022-03-11 兰州交通大学 Method and system for searching rules based on complex network
CN114663245A (en) * 2022-03-16 2022-06-24 南京信息工程大学 Cross-social network identity matching method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150120717A1 (en) * 2013-10-25 2015-04-30 Marketwire L.P. Systems and methods for determining influencers in a social data network and ranking data objects based on influencers
US20160071161A1 (en) * 2014-09-10 2016-03-10 Sysomos L.P. Systems and Methods for Identifying a Target Audience in a Social Data Network
CN108765179A (en) * 2018-04-26 2018-11-06 恒安嘉新(北京)科技股份公司 A kind of credible social networks analysis method calculated based on figure
CN108846422A (en) * 2018-05-28 2018-11-20 中国人民公安大学 Account relating method and system across social networks
CN111309822A (en) * 2020-02-11 2020-06-19 深圳众赢维融科技有限公司 User identity identification method and device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150120717A1 (en) * 2013-10-25 2015-04-30 Marketwire L.P. Systems and methods for determining influencers in a social data network and ranking data objects based on influencers
US20160071161A1 (en) * 2014-09-10 2016-03-10 Sysomos L.P. Systems and Methods for Identifying a Target Audience in a Social Data Network
CN108765179A (en) * 2018-04-26 2018-11-06 恒安嘉新(北京)科技股份公司 A kind of credible social networks analysis method calculated based on figure
CN108846422A (en) * 2018-05-28 2018-11-20 中国人民公安大学 Account relating method and system across social networks
CN111309822A (en) * 2020-02-11 2020-06-19 深圳众赢维融科技有限公司 User identity identification method and device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘苗苗等: "基于共邻节点相似度的加权网络社区发现方法", 《四川大学学报(自然科学版)》 *
张继东等: "基于社区划分和用户相似度的好友信息服务推荐研究", 《情报理论与实践》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112800345A (en) * 2021-02-03 2021-05-14 安徽大学 Community role-aware user demand active prediction method and system
CN112800345B (en) * 2021-02-03 2022-09-30 安徽大学 Community role-aware user demand active prediction method and system
CN114168733A (en) * 2021-12-06 2022-03-11 兰州交通大学 Method and system for searching rules based on complex network
CN114168733B (en) * 2021-12-06 2024-05-24 兰州交通大学 Rule retrieval method and system based on complex network
CN114663245A (en) * 2022-03-16 2022-06-24 南京信息工程大学 Cross-social network identity matching method

Also Published As

Publication number Publication date
CN112069416B (en) 2022-09-02

Similar Documents

Publication Publication Date Title
CN112069416B (en) Cross-social network user identity recognition method based on community discovery
Kumar et al. Identifying influential nodes in Social Networks: Neighborhood Coreness based voting approach
CN104995870B (en) Multiple target server arrangement determines method and apparatus
CN110119475B (en) POI recommendation method and system
CN106960044B (en) Time perception personalized POI recommendation method based on tensor decomposition and weighted HITS
Mo et al. Event recommendation in social networks based on reverse random walk and participant scale control
CN106844407A (en) Label network production method and system based on data set correlation
Ferrara et al. The role of strong and weak ties in Facebook: a community structure perspective
Wang et al. Discover community leader in social network with PageRank
Zhan et al. Identification of top-K influential communities in big networks
Dhumal et al. Survey on community detection in online social networks
CN110119478A (en) A kind of item recommendation method based on similarity of a variety of user feedback datas of combination
Shafik et al. Recommendation system comparative analysis: internet of things aided networks
Gu et al. CAMF: context aware matrix factorization for social recommendation
He et al. A hierarchical matrix factorization approach for location-based web service QoS prediction
Lang et al. Efficient online ad serving in a display advertising exchange
Yuan et al. A Multi‐Granularity Backbone Network Extraction Method Based on the Topology Potential
CN114443972A (en) Information recommendation method, device, equipment and storage medium
CN101442466A (en) Superpose network and implementing method
Xie et al. Evaluation Method of IP Geolocation Database Based on City Delay Characteristics
Wang et al. Who spread to whom? Inferring online social networks with user features
Ma et al. HGL_GEO: Finer-grained IPv6 geolocation algorithm based on hypergraph learning
Maiti et al. Detecting influential users using spread of communications
Chen et al. Irlm: inductive representation learning model for personalized poi recommendation
Wagenseller III et al. Community-based location inference in social media using supervised learning approach

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