Applying Cooperative Games with Coalition Structure for Data Clustering
- 7 Downloads
This paper considers a cooperative game in which the distance (or similarity) between some objects (players) can be measured numerically. For this game, a characteristic function is defined so that it takes high values for the coalitions consisting of most close (similar) players in comparison with the players from the other coalitions. Such a function does not satisfy superadditivity, and hence it seems reasonable to introduce the model with coalition structure. Therefore, this game can be treated as a clustering procedure for objects (players). Finally, the existence conditions of a stable coalition structure are established, which allow to perform efficient (crisp) clustering.
Keywordscoalition stable coalition structure clustering Shapley value Aumann—Dreze value ESD value
Unable to display preview. Download preview PDF.
- 1.Ermolin, N.A., Mazalov, V.V., and Pechnikov, A.A., Game-Theoretic Methods for Finding Communities in Academic Web, Tr. SPIIRAN, 2017, no. 55, pp. 237–254.Google Scholar
- 3.Afsar, M., Energy-Efficient Coalition Formation in Sensor Networks: A Game-Theoretic Approach, Computing Research Repository, 2015, abs/1512.08019.Google Scholar
- 4.Avrachenkov, K.E., Kondratev, A.Yu., and Mazalov, V.V., Cooperative Game Theory Approaches for Network Partitioning, Int. Computing and Combinatorics Conf., 2017, LNCS 10392, pp. 591–602.Google Scholar
- 6.Dhamal, S., Bhat, S., Anoop, K.R., and Embar, V.R., Pattern Clustering Using Cooperative Game Theory, Computing Research Repository, 2012, abs/1201.0461, pp. 653–658.Google Scholar