Node clustering degree based overlapping community division method

A technology of overlapping community and aggregation degree, applied in the field of overlapping community division of social network, can solve the problem of non-overlapping node division and so on

Inactive Publication Date: 2017-07-18
SHANGHAI UNIV
View PDF2 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method can only divide the nodes in the network into a community, and cannot divide overlapping nodes.

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
  • Node clustering degree based overlapping community division method
  • Node clustering degree based overlapping community division method
  • Node clustering degree based overlapping community division method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] Embodiment 1: This overlapping community division method based on node aggregation degree, its operation method steps are as follows:

[0027] The first step is to calculate the degree of aggregation between nodes. First, the global influence of nodes is calculated based on the network topology, and the structural correlation between nodes is calculated by using the local neighbors of nodes. Then the attribute information of nodes is extracted, and the attribute correlation between nodes is calculated. Finally, the node's structural correlation and attribute correlation are combined, and parameters are used to adjust the weight of the two, which is used as the aggregation degree of the node.

[0028] The second step is to use the node aggregation degree to construct the similarity matrix and diagonal matrix of the node, construct the corresponding Laplacian matrix through the similarity matrix and diagonal matrix, and calculate the first K eigenvalues ​​and features of...

Embodiment 2

[0030] Embodiment 2: This embodiment is basically the same as Embodiment 1, and the special features are as follows: the construction method of node aggregation degree in the first step is as follows:

[0031] 1) Node link strength based on network topology

[0032] The calculation of the similarity between nodes in the network is mostly based on the local information of the nodes. If two nodes have the same or similar neighbor nodes, they are considered similar. But it does not take into account the global importance of nodes in the network. This paper incorporates global importance when computing node link strength. This paper uses the improved PageRank [16] algorithm to calculate the importance of nodes in the network, and proposes the concept of node influence. If the node's influence is greater, its global importance in the network is higher.

[0033] Definition 1: Node influence: The influence of a node Vi in G= can be calculated by the following formula

[0034]

...

Embodiment 3

[0061] This embodiment is basically the same as Embodiment 2, and the special features are as follows:

[0062] In order to prove that the overlapping community division method based on the node aggregation degree can indeed have a good effect on overlapping community division, this patent has conducted experiments on the DBLP data set. The DBLP data set is a data set on the scientific research literature cooperation network containing 5,000 authors, which are extracted from 4 research fields of DBLP scientific research literature: dataBase (DB), datamining (DM), information retrieval (IR) and artificial intelligence (AI) . Each author has two attributes: publication volume and main topic

[0063] In Table 2, the modularity value of the algorithm of the present invention is higher than that of COPRA and smaller than that of LFM. However, the obtained Entropy value is smaller than that of other algorithms, and the value of COPARA algorithm fluctuates greatly on different data...

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 node clustering degree based overlapping community division method. The method includes operation steps of calculating the clustering degree of nodes; utilizing the node clustering degree, constructing a characteristic vector matrix; performing community division of the overlapping nodes. According to the invention, a network typological structure and node attributes are integrated and a node clustering degree based overlapping community division algorithm is designed. Based on the overlapping community division algorithm, spectral clustering is combined for overlapping community division. Test results show that the scheme provided by the invention is good in performance and division effect.

Description

technical field [0001] The invention relates to a method for dividing overlapping communities in a social network, in particular to a method for dividing overlapping communities based on node aggregation degree. Background technique [0002] With the rapid development of network technology and the Internet, many online social networks such as Facebook, Twitter, WeChat, etc. are becoming more and more popular. Analyzing social network nodes, the relationship between nodes, and the network structure have attracted the attention of many fields. Community discovery originated from the problem of graph segmentation. Image segmentation is to divide the graph into K subgraphs that are not connected to each other, but the community partition can be divided into any number. Researchers have found that community structures exist in real networks, and discovering communities in networks is of great significance for understanding network structures and analyzing network characteristics...

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): G06Q50/00G06K9/62
CPCG06Q50/01G06F18/23213
Inventor 李卫民蒋署刘炜张礼名
Owner SHANGHAI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap