Using Coalitional Games to Detect Communities in Social Networks
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.
Keywordssocial networks community detection game theory
Unable to display preview. Download preview PDF.
- 5.Lemke, C., Howson, J.: Equilibrium points of bimatrix games. Journal of the Society of Industrial and Applied Mathematics, 413–424 (1964)Google Scholar
- 8.Liu, W.Y., et al.: Intelligent Data Analysis. Science Press, Beijing (2007)Google Scholar
- 12.Shapley, L.S.: A Value for N-person games. In: Kuhn, H.W., Tucker, A.W. (eds.) Contributions to be Theory of Games, pp. 307–317. Princeton University Press (1953)Google Scholar
- 14.Zachary, W.W.: An information flow model for conflict and fission in small groups. Journal of Anthropological Research 33(4), 452–473 (1977)Google Scholar
- 16.Zlotkin, G., Rosenschein, J.: Coalition cryptography and stability mechanisms for coalition formation in task oriented domains. In: Association for the Advancement of Artificial Intelligence, pp. 432–437 (1994)Google Scholar