Semi-supervised community discovery method fusing node attributes and graph structure

Abstract

The invention discloses a semi-supervised community discovery method fusing node attributes and a graph structure, and belongs to the technical field of network analysis. The method comprises the following steps: 1) calculating information entropies of m attributes; 2) calculating attribute similarity; 3) calculating the structural similarity by utilizing the Jaccard similarity; 4) calculating thetotal similarity of the attributes and the structure; 5) searching K initial communities; 6) initializing an initial community matrix; 7) calculating a community division matrix in combination with asemi-supervised method; 8) calculating a reasonable value range of a balance value (trade-off) analysis parameter; 9) obtaining an optimal modularity and a community discovery result according to thetrade-off and the modularity. According to the method, a reasonable division mode is obtained by continuously adjusting parameters involved in the algorithm, and finally, an optimal result of community discovery and a reasonable range of algorithm parameters are given; attributes are fused for community discovery, the reasonable range of the proportion of the attributes is given, and the community discovery modularity and compactness are improved.

Description

technical field [0001] The invention belongs to the technical field of network analysis, and in particular relates to a semi-supervised community discovery method which integrates node attributes and graph structures. Background technique [0002] In recent years, machine learning and data mining have become more popular research directions. The learning and mining of the network will be conducive to the rational use of network data. Networks exist in various fields and are widely used. The social relationship between people can be expressed as a social network, the connection of computers is also a computer network, and the interaction between proteins is also a protein network. And how to reasonably analyze and use these network data has become particularly important. Dividing the Weibo user network into communities is beneficial for companies to recommend friends to users. Dividing the protein interaction network (PPI) can effectively identify key proteins, which is a ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a semi-supervised community discovery method that integrates node attributes and graph structures for the problem that existing social network division methods rarely consider node attributes

Images