A new vertex similarity metric for community discovery: A distance neighbor model

Abstract

The hierarchical clustering methods based on vertex similarity can be employed for community discovery. Vertex similarity metric is the most important part of these methods. However, the existing metrics do not perform well compared with the state-of-the-art algorithms. In this paper, we propose a new vertex similarity metric based on distance neighbor model, called Distance Neighbor Ratio Metric (DNRM), for community discovery. DNRM considers both distance and nearby edge density which are essential measures in community structure. Compared with the existing metrics of vertex similarity, DNRM outperforms substantially in community discovery quality and the computing time. The experiments are designed rigorously involving both well-known social networks in real world and computer generated networks. © 2011 Springer Berlin Heidelberg.