Identifying Clusters with Attribute Homogeneity and Similar Connectivity in Information Networks

With the rapid emergence of the internet world, a lot of information networks become available every day. In many cases, these information networks contain objects connected by multiple links and described by different attributes. In this paper the problem of clustering homogeneous information networks in groups with similar attributes and connections is studied. Clustering such networks is a challenging task due to different importance of links and attributes. In addition, it is not straightforward how to balance the links and attributes information. In this article we describe these challenges and propose a fuzzy clustering model as well as a fuzzy clustering algorithm, HASCOP. Extensive experimentation on real world datasets has shown that HASCOP can be successfully applied in such networks, demonstrating its efficacy and superiority against the state-of-the-art attributed graph clustering methods.