Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Large-scale complex network oriented dense overlapping community division method

A technology of complex networks and overlapping communities, applied in the field of dense and overlapping community division, it can solve the problems of complex computing, loss of useful information, and difficulty in estimating the number of large networks, and achieve the effect of low algorithm complexity.

Inactive Publication Date: 2018-09-14
SUN YAT SEN UNIV
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, there are still many challenges and deficiencies in the current community mining research.
The first is to solve the computational complexity problem of large-scale networks. Existing researches believe that mining high-cohesion and low-coupling community circles in huge social networks requires the design of algorithms with low complexity. Therefore, it is necessary to design Good heuristic algorithm; second, in the mining research of dense overlapping communities, existing research often assumes that the number of mined communities is already known, but this number is generally difficult to estimate and can only be predicted based on experience. Third, in the existing research, researchers often cut the large-scale network abstract graph into several small graphs, which will lose the information of the original graph, although some studies have explored how to Minimize the loss of information in the graph and cut it into several separate graphs, but the useful information of the original graph will always be lost

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Large-scale complex network oriented dense overlapping community division method
  • Large-scale complex network oriented dense overlapping community division method
  • Large-scale complex network oriented dense overlapping community division method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0018] First, use G=(V, E) to abstractly represent a network, which is composed of undirected edges, where V represents the combination of nodes in the network, and E represents the set of edges in the network. If two nodes u, v ∈ V, if there is an edge between these two nodes, then it can be expressed as (u, v) ∈ E. Using ωuv to represent the weight between nodes, the present invention mainly considers the undirected graph, so w uv =1.

[0019] The present invention uses S to represent a collection of several nodes, and defines m S Indicates the weight of the edge in S, then:

[0020]

[0021] Define the weight of the cutting edge of S as c S , then c S defined as:

[0022]

[0023] Define N(S) as the set of neighbor nodes of S. For a neighbor node u∈N(S), the outgoing weight of node u can be defined as:

[0024]

[0025] And the inward weight of node u is:

[0026]

[0027] In the existing community mining, there are many indicators, among which Conductanc...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a large-scale complex network oriented dense overlapping community division method. The method comprises the following steps: S1, abstracting a large network into an undirected graph, initializing system parameters, setting initial community circle set as a null set, and setting iteration stop conditions; S2, traversing nodes that are not accessed, setting the nodes as seeds of community circles, and expanding the circle from the nodes; maintaining an extensible node set of the nodes by using a priority queue, maintaining neighbor nodes by the priority queue, and continuously updating; S3, continuously extending the nodes, completing community extension when conductance of the community circles cannot be reduced any more or reaches the iteration stop conditions, and adding the community circles to the community circle set; S4, judging whether all the nodes are completely accessed or not, and if not, jumping to the S2, otherwise entering S5; S5, outputting the community circle set, and ending the algorithm.

Description

technical field [0001] The invention relates to the fields of parallel computing and social networks, and more specifically, relates to a method for dividing dense and overlapping communities for large-scale complex networks. Background technique [0002] In recent years, with the rise of large social networks (such as Facebook, Twitter and Weibo), community detection (Community Detection) has gradually attracted widespread attention from academia and industry. The goal of community mining is to dig out social circles with high cohesion and low coupling from the abstract network. These circles may be dense and overlapping. At present, Facebook has more than 2 billion monthly active users, so many users have built a huge social network. Community mining can bring a lot of benefits. For example, Facebook can recommend friends to members in circles with the same hobbies; Amazon can mine circles with similar purchasing interests based on its network of goods and users, and reco...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/06G06Q50/00G06F17/30
CPCG06Q10/0631G06Q50/01
Inventor 吴迪叶国桥吴展鹏陈润源
Owner SUN YAT SEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products