Modularity based community detection in heterogeneous networks

Heterogeneous networks consist of different types of nodes and multiple types of edges linking such nodes. While numerous community detection techniques exist for analyzing networks that contain only one type of node, very few such techniques have been developed for heterogeneous networks. Therefore, we propose a modularity-based community detection framework for heterogeneous networks. Unlike existing methods, the proposed approach has the flexibility of treating the number of communities as an unknown quantity. We describe a Louvain-type maximization method for determining the community structure that maximizes the modularity function. Our simulation results show the advantages of the proposed method over the existing methods. Moreover, the proposed modularity function is shown to be consistent under a heterogeneous stochastic blockmodel framework. Analyses of a DBLP four-area data set and a MovieLens data set demonstrate the usefulness of the proposed method.