Enhancing Clustering Methods by Inserting Relational Information

We consider the problem of clustering el-ements that have both content and rela-tional information (e.g. Web-pages, sci-entific papers, etc.). Standard cluster-ing methods use content information only, while graph clustering methods are usu-ally based on the graph structure. Rela-tively recently, researchers have proposed to combine both types of information. In this paper we propose a simple, yet hith-erto unexplored, method to insert the rela-tional information into standard clustering methods.