Community detection based on local central vertices of complex networks

Community structure identification has attracted considerable attention in recent years and there has been many algorithms proposed to detect community structures in complex networks, where some of the algorithms need priori knowledge about number of communities and start to detect the communities by choosing some nodes randomly. This paper proposes a global community detection algorithm based on local central vertices of complex networks. Local degree central vertices are used as initial vertices, by agglomerating the neighbor nodes to local degree central vertices, the community structure can be detected. The experiment results show that our algorithm works effectively and outperforms some other algorithms in terms of modularity.