Community Detection Method Based on Two-layer Dissimilarity of Central Node

Abstract

Studying community discovery algorithms for complex networks is necessary to determine the origin of opinions, analyze the mechanisms of public opinion transmission, and control the evolution of public opinion. The problem of the existing clustering algorithm of the central node having a low quality of community detection must also be solved. This study proposes a community detection method based on the two-layer dissimilarity of the central node (TDCN-CD). First, the algorithm selects the central node through the degree and distance of the node. Selecting nodes in the same community as the central node at the same time is avoided. Simultaneously, the algorithm proposes the dissimilarity index of nodes based on two layers, which can deeply explore the heterogeneity of nodes and achieve the effect of accurate community division. The results of using Karate and Dolphins datasets for simulation show that compared to the Girvan–Newman and Fast–Newman classical community partitioning algorithms, the TDCN-CD algorithm can effectively detect the community structure and more accurately divide the community.