Dynamic community detection based on game theory in social networks

The difficulty of analysis in social networks mainly originates from its huge scale and complicated relations in the network. Social network is a complex system that we can hardly inspect its deep mechanism with only regarding it as a whole. Game theory is a study that focuses on interactions and reactions between intelligent individuals. Recently, game theory based methods are proposed to tackle different kinds of problems in social networks. In this paper, we focus on a tough problem that detect communities in dynamic networks. We present a game theoretic approach for community detection in dynamic networks, in which each node regarded as an intelligent and selfish agent chooses from a set of actions, that is, join, leave or switch a community in order to maximize its utility. The experimental results show the effectiveness and the advantages of our model.