CN106055568B - A kind of automatic group technology of friend of the social networks based on single step addition group - Google Patents
A kind of automatic group technology of friend of the social networks based on single step addition group Download PDFInfo
- Publication number
- CN106055568B CN106055568B CN201610338452.XA CN201610338452A CN106055568B CN 106055568 B CN106055568 B CN 106055568B CN 201610338452 A CN201610338452 A CN 201610338452A CN 106055568 B CN106055568 B CN 106055568B
- Authority
- CN
- China
- Prior art keywords
- user
- group
- corporations
- times iteration
- social networks
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The invention discloses a kind of automatic group technology of social networks friend based on single step addition group, social networks is described as a binary group by this method, solves the automatic grouping problem of social networks friend using based on the overlapping corporations detection algorithm of single step addition group.The present invention can fast implement circle of friends in social networks and be grouped automatically, improve grouping efficiency and accuracy, so as to be grouped by accurate friend, recommend more reliably friend for user, reduce the unnecessary trouble that user generates when searching for friend with a common goal.
Description
Technical field
It is specifically exactly to propose one kind based on single step the present invention relates to the automatic group technology field of friend in social networks
The automatic group technology of friend of the social networks of the group of addition.
Background technique
With the arrival of Internet era, social networks, which springs up like bamboo shoots after a spring rain, to be developed by leaps and bounds, and has attracted the country
The extensive concern of outer scholar.The appearance of social networks changes the traditional thinking of people, exchange, work and life style, more next
More multi-user selects that the exchange that online social networks carries out information is added, and the userbase of social networks is in explosive growth, is added
The speed that information increases in acute network.Needed for the aggravation of the network information causes user that can not select oneself from the data of magnanimity
Information, that is, user faces problem of information overload.If user carries out manually the online social networks where oneself
Friend's grouping, it is not only time-consuming also laborious in this way, cause many users not classify to their friend, it is therefore, most of
Social network sites all can be user classification good friend according to certain public attributes of user automatically.
The automatic grouping problem in social networks friend recommendation includes two class research methods at present:
A kind of research method is the research method based on social networks global characteristics, and such methods need to detect social networks
In all link structure, know the global information of network.But such methods calculate higher cost, in complicated social networks
The middle high efficiency that may will affect proposed algorithm.
A kind of research method is based on social networks partial structurtes characteristic, such method is to grind with the structure of nodes
Study carefully friend's grouping problem in object implementatio8 friend recommendation.Such methods computation complexity is lower, but is typically due to letter
Ceasing insufficient causes in social networks friend to be grouped inaccuracy.
Summary of the invention
The present invention is directed to the shortcomings of the prior art, provides a kind of social networks based on single step addition group
The automatic group technology of friend is grouped automatically to fast implement circle of friends in social networks, improves grouping efficiency and accuracy,
So as to be grouped by accurate friend, recommend more reliably friend for user, reduces user and searching for friend with a common goal
When the unnecessary trouble that generates.
The present invention adopts the following technical scheme that in order to solve the technical problem
A kind of the characteristics of friend's automatic group technology of social networks based on single step addition group of the present invention is by following step
It is rapid to carry out:
Step 1: defining the social networks is characterized as binary group { V, E }, V={ v1,v2,…,vi,…,vnIndicate institute
State the set of all users in social networks, viIndicate i-th of user;N is the sum of user;E={ eij| i=1,2 ..., n;j
=1,2 ..., n indicate any two user between connection set;eijIndicate i-th of user viWith j-th of user vjIt
Between connection;If eij=1 indicates i-th of user viWith j-th of user vjBetween have Bian Xianglian, and i-th of user viWith j-th
User vjIt is known as neighbor user each other;If eij=0, indicate i-th of user viWith j-th of user vjBetween it is boundless be connected, i.e., do not deposit
It is contacting;
Step 2: the social networks G is divided into X using the overlapping network corporations detection algorithm based on single step addition group
A grouping set is denoted as C={ C1,C2,…,Cx,…,CX};CxIndicate x-th of grouping;X=1,2 ..., X;To realize social activity
The friend of network is grouped automatically.
The characteristics of friend's automatic group technology of social networks of the present invention based on single step addition group, lies also in,
The overlapping network corporations detection algorithm based on single step addition group in the step 2 is to carry out as follows:
Step 0, initialization x=1;Using the set V of all users of the social networks as the candidate of x-th of prepared group
User's set, is denoted as V(x);Define iteration variable t;Define iteration variable r;
Step 1, initialization t=1;
Step 2, from the candidate user V of x-th of prepared group(x)In randomly select the user v of the t times iteration(x)(t);
Step 3, the user v that the t times iteration is calculated using formula (1)(x)(t)Cluster coefficients G(x)(t):
In formula (1),Indicate the user v of the t times iteration in x-th of prepared group(x)(t)Neighbor user quantity,Indicate the user v of the t times iteration(x)(t)It is allThe number of edges for having side connected between a neighbor user;
Step 4, the user v that the t times iteration in x-th of prepared group is calculated using formula (1)(x)(t)'sA neighbor user
Cluster coefficients, and respectively with calculate the t times iteration user v(x)(t)Cluster coefficients G(x)(t)Compare, if can find has maximum
Cluster coefficients and maximum cluster coefficients are greater than G(x)(t)Neighbor user, then using the neighbor user found as x-th of prepared group
In the t+1 times iteration user v(x)(t+1);And execute step 5;Otherwise, retain the user v of the t times iteration(x)(t)As
Core customer in x prepared group;
T+1 is assigned to t by step 5;And repeat step 3;
Step 6, initialization r=1;
Step 7, the core customer v for finding the t times iteration in x-th of prepared group(x)(t)The k group connected;The core
Heart user v(x)(t)All users in k group connected are neighbor user;With the core customer v(x)(t)The k group connected
Corporations C as the r times iteration in x-th of prepared group(x)(r);k≥3;
Step 8, the corporations C that the r times iteration in x-th of prepared group is calculated according to formula (2)(x)(r)Corporations fitness value F(x)(r):
In formula (2), α is adjustable parameter;Respectively indicate the initial corporations C of the r times iteration(x)(r)'s
Internal degree and external degree, internal degreeFor the corporations C of the r times iteration(x)(r)In there is between all users side to be connected
2 times of number of edges;External degreeFor the corporations C of the r times iteration(x)(r)In all users and external user have while be connected while
Number;
Step 9, the corporations C for finding the r times iteration(x)(r)In all users neighbor user and carry out union processing, thus
Constitute the corporations C of the r times iteration in x-th of prepared group(x)(r)Neighbor user;
Step 10, the corporations C for judging the r times iteration in x-th of prepared group(x)(r)Neighbor user respectively whether deposit
In the k group itself connected;If it exists, then respective the connected k group of neighbor user is added separately to the r times
The corporations C of iteration(x)(r)It is middle to calculate corresponding corporations' fitness value;If it does not exist, then directly corresponding neighbor user is added to
The corporations C of the r times iteration(x)(r)It is middle to calculate corresponding corporations' fitness value;
If step 11 can be found so that corporations' fitness value F(x)(r)Rise in value maximum neighbor user or neighbor user
The k group connected;The k group that rise in value maximum neighbor user or neighbor user are connected then is added to the initial of the r times iteration
Corporations C(x)(r)In, and the corporations C as the r+1 times iteration(x)(r+1), execute step 12;Otherwise, it protects in x-th of prepared group
The initial corporations C of the r times iteration(x)(r)It is grouped as x-th;And execute step 13;
R+1 is assigned to r by step 12;And repeat step 8;
Step 13, from the candidate user V of x-th of prepared group(x)It is middle removal it is described x-th grouping, thus obtain xth+
The candidate user V of 1 prepared group(x+1);
X+1 is assigned to x by step 14;And return step 1 executes, to obtain X grouping, and then realizes social networks
Friend be grouped automatically.
Compared with the prior art, the invention has the advantages that:
1, the thought that the present invention expands according to local corporations is based on use when friend is grouped automatically to social networks
Corporations' extended mode of single step addition group, to complete the automatic grouping of social networks.The present invention expands process in corporations and only needs
The local message for obtaining user in social networks can complete user and be grouped automatically, without knowing the overall situation of entire social networks
Information and nodal information can quickly detect each friend grouping of social networks in this way, obtain in social networks more
Accurate friend's group result;It solves the problems, such as to calculate cost based on global structure characterization method height, also solves certain bases
Network divides the problem of inaccuracy in partial structurtes characterization method.
2, it first has to that initial friend is selected to be grouped when carrying out friend and being grouped automatically, initial friend grouping of the invention is logical
Cross what core node was searched for.Core node is the centromere of grouping as node most special in a grouping or figure
Point simultaneously plays key effect to the connection of user in being grouped, can avoid being selected by initial packet so to a certain extent inaccurate
True bring randomness, to improve the accuracy that social networks friend is grouped automatically.
3, presently, there are based on the method locally expanded in friend's grouping process every time only be added a user node,
Do not fully consider that the local message of user node will lead to grouping inaccuracy.The present invention quotes the concept of group: group is one complete
Full-mesh subgraph, they very likely belong to a grouping in grouping partition process, so an if user node category
In some grouping, then the group where the user node also more likely becomes a part of this grouping.The present invention is expanding
A group is added when filling every time, not only allow for being added user's point and has the bonding strength of grouping, while preferably being considered
The connection being added inside user, more can accurately find to belong to while the user of multiple groupings.
Detailed description of the invention
Fig. 1 is inventive algorithm flow chart;
Fig. 2 is the simple network schematic diagram that the present invention illustrates.
Specific embodiment
In embodiment, a kind of automatic group technology of social networks friend based on single step addition group will be in social networks
The automatic grouping problem of friend is converted into complex network community test problems, by utilizing the complex network society based on single step addition group
Detection algorithm is rolled into a ball to solve the automatic grouping problem of friend in social networks, to obtain friend's grouping in social networks;Specifically
It is to carry out as follows that ground, which is said:
Step 1: defining the social networks is characterized as binary group { V, E }, V={ v1,v2,…,vi,…,vnIndicate institute
State the set of all users in social networks, viIndicate i-th of user;N is the sum of user;E={ eij| i=1,2 ..., n;j
=1,2 ..., n indicate any two user between connection set;eijIndicate i-th of user viWith j-th of user vjIt
Between connection;If eij=1 indicates i-th of user viWith j-th of user vjBetween have Bian Xianglian, and i-th of user viWith j-th
User vjIt is known as neighbor user each other;If eij=0, indicate i-th of user viWith j-th of user vjBetween it is boundless be connected, i.e., do not deposit
It is contacting;It is illustrated in figure 2 the simple social network structure figure comprising 11 users, wherein each node on behalf social network
User in network, each edge, which represents, has connection between user;
Step 2: the social networks G is divided into X using the overlapping network corporations detection algorithm based on single step addition group
A grouping set is denoted as C={ C1,C2,…,Cx,…,CX};CxIndicate x-th of grouping;X=1,2 ..., X;To realize social activity
The friend of network is grouped automatically.
Specifically, as shown in Figure 1, based on single step addition group overlapping network corporations detection algorithm be as follows into
Row:
Step 0, initialization x=1;Using the set V of all users of the social networks as the candidate of x-th of prepared group
User's set, is denoted as V(x);Define iteration variable t;Define iteration variable r;
Step 1, initialization t=1;
Step 2, from the candidate user V of x-th of prepared group(x)In randomly select the user v of the t times iteration(x)(t);
Step 3, the user v that the t times iteration is calculated using formula (1)(x)(t)Cluster coefficients G(x)(t):
In formula (1),Indicate the user v of the t times iteration in x-th of prepared group(x)(t)Neighbor user quantity,Indicate the user v of the t times iteration(x)(t)It is allThe number of edges for having side connected between a neighbor user;
Step 4, the user v that the t times iteration in x-th of prepared group is calculated using formula (1)(x)(t)'sA neighbor user
Cluster coefficients, and respectively with calculate the t times iteration user v(x)(t)Cluster coefficients G(x)(t)Compare, if can find has maximum
Cluster coefficients and maximum cluster coefficients are greater than G(x)(t)Neighbor user, then using the neighbor user found as x-th of prepared group
In the t+1 times iteration user v(x)(t+1);And execute step 5;Otherwise, retain the user v of the t times iteration(x)(t)As
Core customer in x prepared group;
T+1 is assigned to t by step 5;And repeat step 3;
Step 6, initialization r=1;
Step 7, the core customer v for finding the t times iteration in x-th of prepared group(x)(t)The k group connected;The core
Heart user v(x)(t)All users in k group connected are neighbor user;With the core customer v(x)(t)The k group connected
Corporations C as the r times iteration in x-th of prepared group(x)(r);k≥3;Such as in FIG. 2, it is assumed that preliminary preparatory group is saved by user
Point { 1,2,3,4,5 } is constituted, at this point, setting k=3;
Step 8, the corporations C that the r times iteration in x-th of prepared group is calculated according to formula (2)(x)(r)Corporations fitness value F(x)(r):
In formula (2), α is adjustable parameter;Respectively indicate the initial corporations C of the r times iteration(x)(r)'s
Internal degree and external degree, internal degreeFor the corporations C of the r times iteration(x)(r)In there is between all users side to be connected
2 times of number of edges;External degreeFor the corporations C of the r times iteration(x)(r)In all users and external user have while be connected while
Number is 0.824 according to the fitness value that formula (2) calculates the 1st preparation grouping { 1,2,3,4,5 };
Step 9, the corporations C for finding the r times iteration(x)(r)In all users neighbor user and carry out union processing, thus
Constitute the corporations C of the r times iteration in x-th of prepared group(x)(r)Neighbor user, in this way known to the 1st preparation grouping 1,2,3,
4,5 } neighbor user is { 6,7,8 };
Step 10, the corporations C for judging the r times iteration in x-th of prepared group(x)(r)Neighbor user respectively whether deposit
In the k group itself connected;If it exists, then respective the connected k group of neighbor user is added separately to the r times
The corporations C of iteration(x)(r)It is middle to calculate corresponding corporations' fitness value;If it does not exist, then directly corresponding neighbor user is added to
The corporations C of the r times iteration(x)(r)It is middle to calculate corresponding corporations' fitness value;Then in the present embodiment, the 1st preparation grouping 1,2,
3,4,5 } the k group that neighbor user { 6,7,8 } is connected is respectively { (6,7,8), (6,7,8), (8,9,11) };
If step 11 can be found so that corporations' fitness value F(x)(r)Rise in value maximum neighbor user or neighbor user
The k group connected;The k group that rise in value maximum neighbor user or neighbor user are connected then is added to the initial of the r times iteration
Corporations C(x)(r)In, and the corporations C as the r+1 times iteration(x)(r+1), execute step 12;Otherwise, it protects in x-th of prepared group
The initial corporations C of the r times iteration(x)(r)It is grouped as x-th;And execute step 13;In the present embodiment, No. 6 are known by calculating
The fitness value for the grouping { 1,2,3,4,5,6,7,8 } that the 1st prepared group { 1,2,3,4,5 } obtains is added in k group where user
It is 0.897, so that the fitness value increase of grouping is maximum, therefore the k group of No. 6 users' connections of selection is added in prepared grouping and obtains
To new grouping { 1,2,3,4,5,6,7,8 }, if the suitable of grouping will not be made by adding the k group that new user or user are connected again
The 1st for answering angle value to increase, therefore obtaining this social networks is grouped into { 1,2,3,4,5,6,7,8 };
R+1 is assigned to r by step 12;And repeat step 8;
Step 13, from the candidate user V of x-th of prepared group(x)It is middle removal it is described x-th grouping, thus obtain xth+
The candidate user V of 1 prepared group(x+1), at this point, candidate user is { 9,10,11,12 } in Fig. 2 example;
X+1 is assigned to x by step 14;And return step 1 executes, to obtain X grouping, and then realizes social networks
Friend be grouped automatically, in the example that Fig. 2 is shown, be grouped into automatically using the obtained friend of the present invention 1,2,3,4,5,6,
7,8 } and { 9,10,11,12 }.
Claims (1)
1. a kind of automatic group technology of friend of the social networks based on single step addition group, it is characterized in that carrying out as follows:
Step 1: defining the social networks is characterized as binary group { V, E }, V={ v1,v2,…,vi,…,vnIndicate the social activity
The set of all users, v in networkiIndicate i-th of user;N is the sum of user;E={ eij| i=1,2 ..., n;J=1,
2 ..., n indicate any two user between connection set;eijIndicate i-th of user viWith j-th of user vjBetween
Connection;If eij=1 indicates i-th of user viWith j-th of user vjBetween have Bian Xianglian, and i-th of user viWith j-th of user
vjIt is known as neighbor user each other;If eij=0, indicate i-th of user viWith j-th of user vjBetween it is boundless be connected, i.e., there is no connection
System;
Step 2: the social networks G is divided into X points using the overlapping network corporations detection algorithm based on single step addition group
Group set, is denoted as C={ C1,C2,…,Cx,…,CX};CxIndicate x-th of grouping;X=1,2 ..., X;To realize social networks
Friend be grouped automatically;
The overlapping network corporations detection algorithm based on single step addition group in the step 2 is to carry out as follows:
Step 0, initialization x=1;Using the set V of all users of the social networks as the candidate user of x-th of prepared group
Set, is denoted as V(x);Define iteration variable t;Define iteration variable r;
Step 1, initialization t=1;
Step 2, from the candidate user V of x-th of prepared group(x)In randomly select the user v of the t times iteration(x)(t);
Step 3, the user v that the t times iteration is calculated using formula (1)(x)(t)Cluster coefficients G(x)(t):
In formula (1),Indicate the user v of the t times iteration in x-th of prepared group(x)(t)Neighbor user quantity,Table
Show the user v of the t times iteration(x)(t)It is allThe number of edges for having side connected between a neighbor user;
Step 4, the user v that the t times iteration in x-th of prepared group is calculated using formula (1)(x)(t)'sA neighbor user gathers
Class coefficient, and the user v with the t times iteration of calculation respectively(x)(t)Cluster coefficients G(x)(t)Compare, if can find has maximum cluster
Coefficient and maximum cluster coefficients are greater than G(x)(t)Neighbor user, then using the neighbor user found as in x-th of prepared group
The user v of t+1 iteration(x)(t+1);And execute step 5;Otherwise, retain the user v of the t times iteration(x)(t)As x-th
Core customer in prepared group;
T+1 is assigned to t by step 5;And repeat step 3;
Step 6, initialization r=1;
Step 7, the core customer v for finding the t times iteration in x-th of prepared group(x)(t)The k group connected;The core is used
Family v(x)(t)All users in k group connected are neighbor user;With the core customer v(x)(t)The k group conduct connected
The corporations C of the r times iteration in x-th of prepared group(x)(r);k≥3;
Step 8, the corporations C that the r times iteration in x-th of prepared group is calculated according to formula (2)(x)(r)Corporations fitness value F(x)(r):
In formula (2), α is adjustable parameter;Respectively indicate the initial corporations C of the r times iteration(x)(r)Inside degree
Several and external degree, internal degreeFor the corporations C of the r times iteration(x)(r)In have between all users side connected number of edges 2
Times;External degreeFor the corporations C of the r times iteration(x)(r)In all users number of edges for thering is side to be connected with external user;
Step 9, the corporations C for finding the r times iteration(x)(r)In all users neighbor user and carry out union processing, to constitute
The corporations C of the r times iteration in x-th of prepared group(x)(r)Neighbor user;
Step 10, the corporations C for judging the r times iteration in x-th of prepared group(x)(r)Neighbor user respectively whether there is from
The k group that body is connected;If it exists, then respective the connected k group of neighbor user is added separately to the r times iteration
Corporations C(x)(r)It is middle to calculate corresponding corporations' fitness value;If it does not exist, then corresponding neighbor user is directly added to r
The corporations C of secondary iteration(x)(r)It is middle to calculate corresponding corporations' fitness value;
If step 11 can be found so that corporations' fitness value F(x)(r)Rise in value maximum neighbor user or company, neighbor user institute
The k group connect;The k group that rise in value maximum neighbor user or neighbor user are connected then is added to the initial corporations C of the r times iteration(x)(r)In, and the corporations C as the r+1 times iteration(x)(r+1), execute step 12;Otherwise, it protects in x-th of prepared group the r times
The initial corporations C of iteration(x)(r)It is grouped as x-th;And execute step 13;
R+1 is assigned to r by step 12;And repeat step 8;
Step 13, from the candidate user V of x-th of prepared group(x)Middle removal x-th of grouping, to obtain (x+1)th
The candidate user V of prepared group(x+1);
X+1 is assigned to x by step 14;And return step 1 executes, to obtain X grouping, and then realizes the friend of social networks
Friendly automatic grouping.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610338452.XA CN106055568B (en) | 2016-05-18 | 2016-05-18 | A kind of automatic group technology of friend of the social networks based on single step addition group |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610338452.XA CN106055568B (en) | 2016-05-18 | 2016-05-18 | A kind of automatic group technology of friend of the social networks based on single step addition group |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106055568A CN106055568A (en) | 2016-10-26 |
CN106055568B true CN106055568B (en) | 2019-06-28 |
Family
ID=57176615
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610338452.XA Active CN106055568B (en) | 2016-05-18 | 2016-05-18 | A kind of automatic group technology of friend of the social networks based on single step addition group |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106055568B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109978705A (en) * | 2019-02-26 | 2019-07-05 | 华中科技大学 | Combo discovering method in a kind of social networks enumerated based on Maximum Clique |
CN110287237B (en) * | 2019-06-25 | 2021-07-09 | 上海诚数信息科技有限公司 | Social network structure analysis based community data mining method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101344940A (en) * | 2008-08-21 | 2009-01-14 | 魏芳 | Network overlapped corporation detection method based on global partition and local expansion |
CN101887573A (en) * | 2010-06-11 | 2010-11-17 | 北京邮电大学 | Social network clustering correlation analysis method and system based on core point |
CN102611588A (en) * | 2012-03-28 | 2012-07-25 | 西安电子科技大学 | Method for detecting overlapped community network based on automatic phase conversion clustering |
CN103345531A (en) * | 2013-07-26 | 2013-10-09 | 苏州大学 | Method and device for determining network community in complex network |
CN103400299A (en) * | 2013-07-02 | 2013-11-20 | 西安交通大学 | Method for detecting network overlapped communities based on overlapped point identification |
CN104021233A (en) * | 2014-06-30 | 2014-09-03 | 电子科技大学 | Social network friend recommendation method based on community discovery |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050209999A1 (en) * | 2004-03-19 | 2005-09-22 | Kenny Jou | Systems and methods for class designation in a computerized social network application |
CN101227422B (en) * | 2007-11-26 | 2012-01-25 | 腾讯科技(深圳)有限公司 | Method and system for locating linkman locality grouping |
CN102347917B (en) * | 2011-11-04 | 2014-04-16 | 西安电子科技大学 | Contact semantic grouping method for network message communication |
CN102662964B (en) * | 2012-03-05 | 2015-05-27 | 北京千橡网景科技发展有限公司 | Method and device for grouping friends of user |
CN103426042B (en) * | 2012-05-15 | 2016-04-13 | 腾讯科技(深圳)有限公司 | The group technology of social networks and system |
CN102930362A (en) * | 2012-09-17 | 2013-02-13 | 百度在线网络技术(北京)有限公司 | Method and device for grouping contact persons in address list and mobile terminal |
WO2014176772A1 (en) * | 2013-05-02 | 2014-11-06 | 东莞宇龙通信科技有限公司 | Terminal and method for grouping contact persons |
CN103870547A (en) * | 2014-02-26 | 2014-06-18 | 华为技术有限公司 | Grouping processing method and device of contact persons |
CN104199964B (en) * | 2014-09-19 | 2017-12-12 | 大连民族学院 | Information processing method and device |
CN105045833B (en) * | 2015-06-30 | 2018-12-28 | 北京嘀嘀无限科技发展有限公司 | The classification method and device of user's friend relation |
CN104731962B (en) * | 2015-04-03 | 2018-10-12 | 重庆邮电大学 | Friend recommendation method and system based on similar corporations in a kind of social networks |
-
2016
- 2016-05-18 CN CN201610338452.XA patent/CN106055568B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101344940A (en) * | 2008-08-21 | 2009-01-14 | 魏芳 | Network overlapped corporation detection method based on global partition and local expansion |
CN101887573A (en) * | 2010-06-11 | 2010-11-17 | 北京邮电大学 | Social network clustering correlation analysis method and system based on core point |
CN102611588A (en) * | 2012-03-28 | 2012-07-25 | 西安电子科技大学 | Method for detecting overlapped community network based on automatic phase conversion clustering |
CN103400299A (en) * | 2013-07-02 | 2013-11-20 | 西安交通大学 | Method for detecting network overlapped communities based on overlapped point identification |
CN103345531A (en) * | 2013-07-26 | 2013-10-09 | 苏州大学 | Method and device for determining network community in complex network |
CN104021233A (en) * | 2014-06-30 | 2014-09-03 | 电子科技大学 | Social network friend recommendation method based on community discovery |
Also Published As
Publication number | Publication date |
---|---|
CN106055568A (en) | 2016-10-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102347917B (en) | Contact semantic grouping method for network message communication | |
JP2016503937A (en) | Search and determine activity areas | |
CN103838820B (en) | Evolutionary multi-objective optimization community detection method based on affinity propagation | |
CN103888541A (en) | Method and system for discovering cells fused with topology potential and spectral clustering | |
CN106055568B (en) | A kind of automatic group technology of friend of the social networks based on single step addition group | |
CN105354260A (en) | Mobile application recommendation method with social network and project feature fused | |
CN104102699B (en) | A kind of subgraph search method in the set of graphs that clusters | |
CN109213926A (en) | A kind of location recommendation method divided based on community with Multi-source Information Fusion | |
CN110275929A (en) | A kind of candidate road section screening technique and mesh segmentation method based on mesh segmentation | |
CN107085594A (en) | Subgraph match method based on set similarity in big chart database | |
CN105045967B (en) | Complex network important node sort method based on group's degree and model evolution method | |
CN112069416B (en) | Cross-social network user identity recognition method based on community discovery | |
CN106530100A (en) | Community discovery technical method facing confidence-level social network | |
CN111310849A (en) | Effective link prediction method based on KNN | |
CN108280548A (en) | Intelligent processing method based on network transmission | |
CN105701511B (en) | A kind of Adaptive spectra clustering method extracting network node community attributes | |
CN113034297A (en) | Complex network key node identification method and system based on node attraction | |
CN108984630B (en) | Application method of node importance in complex network in spam webpage detection | |
CN106127595A (en) | A kind of community structure detection method based on positive and negative side information | |
CN112035545B (en) | Competition influence maximization method considering non-active node and community boundary | |
CN108924196A (en) | Industry internet green energy resource management system | |
CN109218184A (en) | Router home AS recognition methods based on port and structural information | |
CN108319727A (en) | A method of any two points shortest path in social networks is found based on community structure | |
CN107423319A (en) | A kind of spam page detection method | |
CN113378339A (en) | Ordering method of social network nodes |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |