Overlapping Community Detection based On Maximal Clique and Multi-objective Ant Colony Optimization

Abstract

Community detection is an important technique of complex network, which has great influence in understanding the structure of networks. There are much attention has been paid to detect separated communities, where one node must belong to one community. However, it is more necessary for us to detect the overlapping communities, which often appear in many real-world networks. In this paper, we proposed a multi-objective overlapping community detection algorithm called MCMOCD-ACO, which combing with the multiobjective ant colony optimization(mACO) and clique-based representation scheme. In this algorithm, a new representation scheme based on the introduced maximal-clique graph is presented. Since the maximal clique graph is defined by using a set of maximal cliques of original graph as nodes and two maximal cliques are allowed to share the same nodes of the original graph, overlap is an intrinsic property of the maximal-clique graph. Under this representation scheme, mACO is allowed to handle the separated community in a way similar to that of the separated community detection which can simplified the overlapping community detection optimization problems. The experimental results on synthetic and real networks demonstrate the effective performance of our algorithm over four state-of-the-art overlapping community methods.