A general probabilistic framework for detecting community structure in networks

Based on Newman's fast algorithm, in this paper we develop a general probabilistic framework for detecting community structure in a network. The key idea of our generalization is to characterize a network (graph) by a bivariate distribution that specifies the probability of the two vertices appearing at both ends of a randomly selected path in the graph. With such a bivariate distribution, we give a probabilistic definition of a community and a definition of a modularity index. To detect communities in a network, we propose a class of distribution-based clustering algorithms that have comparable computational complexity to that of Newman's fast algorithm. Our generalization provides the additional freedom to choose a bivariate distribution and a correlation measure. As such, we obtain significant performance improvement over the original Newman fast algorithm in the computer simulations of random graphs with known community structure.