Using Coalitional Games to Detect Communities in Social Networks

Abstract

The community detection in social networks is important to understand the structural and functional properties of networks. In this paper we propose a coalitional game model for community detection in social networks, and use the Shapley Value in coalitional games to evaluate each individual’s contribution to the closeness of connection. We then develop an iterative formula for computing the Shapley Value to improve the computation efficiency. We further propose a hierarchical clustering algorithm GAMEHC to detect communities in social networks. The effectiveness of our methods is verified by preliminary experimental result.