This paper considers the difficulties in community discovery, and comes up with a community discovery algorithm on the basis of role assorted thoughts. Previous work indicates that a robust approach to community detection is the maximization of inner communication and the minimization of the inout interaction. Here we show that this problem can be solved accords to the role assorted method which give distinguish labels to vertices in the same community. This method leads us to a number of possible algorithms for detecting community structures in both unweighted and weighted networks. The applicability and expandability of algorithms proposed are illustrated with application to a variety of computergenerated networks and realworld complex networks.
展开▼
