Centrality Analysis, Role-Based Clustering, and Egocentric Abstraction for Heterogeneous Social Networks

Abstract

The social network is a powerful data structure allowing the depiction of relationship information between entities. Recent researchers have proposed many successful methods on analyzing homogeneous social networks assuming only a single type of node and relation. Nevertheless, real-world complex networks are usually heterogeneous, which presumes a network can be composed of different types of nodes and relations. In this paper, we propose an unsupervised tensor-based mechanism considering higher-order relational information to model the complex semantics of a heterogeneous social network. Based on the model we present solutions to three critical issues in heterogeneous networks. The first concerns identifying central nodes in the heterogeneous network. Second, we propose a role-based clustering method to identify nodes which play similar roles in the network. Finally, we propose an egocentric abstraction mechanism to facilitate further explorations in a complex social network. The evaluations are conducted on a real-world movie dataset and an artificial crime dataset with promising results.