Community Detection on Heterogeneous Networks

Abstract

Community detection is a popular research topic in network study. It helps us to understand the relationship between entities of a network. In this research, we propose a new method for performing community detection on heterogeneous networks. Previous researches about community detection focus overwhelmingly on homogeneous networks, which are suitable for describing homogeneous systems. While real-world networks are often heterogeneous, in the sense that they are composed of different types of entities and relationships. For example, in social tagging system such as Delicious or Flickr, a network can be composed of nodes of users, tags and documents. Therefore, heterogeneous networks that represent real-world relationships can contain more information than homogeneous networks. In this paper, we propose a method for transforming a heterogeneous network into a bipartite network. Then community detection is performed on the bipartite network, which can be applied on large-scale networks and can handle arbitrary types of nodes and edges. We performed experiments with our method on real-world large-scale dataset.