Recherche et représentation de communautés dans un grand graphe: Une approche combinée

Abstract

This paper deals with the analysis and the visualization of large graphs. The sizes of these graphs become increasingly large as data acquisition and storage are developing. Hence the development of methods for graph analysis and visualization is a dynamic research area. In this paper we propose a method for graphs representation and visualization based on a prior clustering of the vertices. The clustering step consists in optimizing a quality measure specifically suited for the search of dense groups in graphs. The visualization step is based on a constrained force-directed algorithm. Finally the approach is illustrated by two applications coming from the social network field. © 2011 Lavoisier, Paris.