A Micro-Community Structure Merging Model Using a Community Sample Matrix

Detecting an overlapping and hierarchical community structure can give a significant insight into structural and functional properties in complex networks. We propose a micro-community structure merging model to detect overlapping and hierarchical communities. The algorithm maps communities to random variables using the community sample matrix to evaluate similarity between communities. After finding density-based micro-community structures, the algorithm merges these reasonable micro-communities iteratively to form communities. Simulation results in three real networks show that the proposed algorithm is more accurate than some existing mechanisms. In this way, we can obtain a detailed understanding of the overlapping and hierarchical communities.