Overlapping Community Structure and Modular Overlaps in Complex Networks

Abstract

In order to find overlapping community structure of complex networks, many researchers make endeavours. Here, we first discuss some existing functions proposed for measuring the quality of overlapping community structure. Second, we propose a novel algorithm called fuzzy detection for overlapping community detection. Our new method benefits from an existing partition detection technique and aims at identifying modular overlaps. A modular overlap is a group of overlapping nodes. Therefore, the overlaps shared by several communities are possibly grouped into several different modular overlaps. The results in synthetic networks and real networks demonstrate that our method can uncover and characterize meaningful overlapping nodes.