Efficiently mining community structures in oriented social networks

A community within a network is a group of vertices densely connected to each other but less connected to the vertices outside. The problem of detecting communities in directed networks plays a key role in a wide range of research areas, e.g. Computer Science, Biology and Sociology. Most of the existing algorithms to find communities count on the topological features of the network and often do not scale well on directs, real-life instances. In this article, We show how the widely used benefit function that we define Dense_Pur can incorporate the information contained in edge directions. We propose a graph mining algorithm, called ACODIG1. for maximizing our objective function over using Ant Colony Optimisation. Test on an a simple directed netwok, show the efficiency of ACODIG to detecting communities in directed network.