Solving dynamic overlapping community detection problem by a multiobjective evolutionary algorithm based on decomposition

Dynamic and overlapping are two common features of community structures for many real world complex networks. Although there are few studies on detecting dynamic overlapping communities, all those studies only consider a single optimization objective. In practice, it is necessary to evaluate the community detection by multiple metrics to reflect different aspects of a community structure and those metrics may conflict with each other. In this paper, we propose a multi-objective approach based on decomposition for the problem of dynamic overlapping community detection, with consideration of three optimization objectives: partition density (D), extended modularity (EQ), and improved mutual information (NMILFK). The dynamic overlapping network can be regarded as a set of network snapshots. The multi-objective evolutionary algorithm based on decomposition (MOEA/D) is used to detect the communities for each snapshot. To improve the search efficiency, the dynamic optimization technique and a dynamic resource allocation strategy are introduced into the approach. Experiments show that our approach can find uniformly distributed Pareto solutions for the problem and outperforms those comparative approaches.