A game theoretic approach to community detection in social networks

Abstract

The problem of detecting community structures in social networks is a complex problem of great importance in sociology, biology and computer science. Communities are characterized by dense intra-connections and comparatively sparse inter-cluster connections. The community detection problem is empirically formulated from a game theoretic point of view and solved using a Crowding based Differential Evolution algorithm adapted for detecting Nash equilibria of noncooperative games. Numerical results indicate the potential of this approach. © 2011 Springer-Verlag Berlin Heidelberg.