Публикации
Mazalov V.V.
Comparing Game-Theoretic and Maximum Likelihood Approaches for Network Partitioning
// Transactions on Computational Collective Intelligence, XXXI, LNCS 11290. Springer, 2018. Pp. 37-46
Ключевые слова: Network partitioning, Community detection, Cooperative games, Hedonic games, Tuning of the parameter, Maximum likelihood method
The purpose of this article is to show the relationship between the game-theoretic approach and the maximum likelihood method in the problem of community detection in networks. We formulate a cooperative game related with network structure where the nodes are players in a hedonic game and then we find the stable partition of the network into coalitions. This approach corresponds to the problem of maximizing a potential function and allows to detect clusters with various resolution. We propose here the maximum likelihood method for the tuning of the resolution parameter in the hedonic game. We illustrate this approach by some numerical examples.
Последние изменения: 8 января 2019