Overlapping and Hierarchical Community Discovery Based on Spectral Method

Abstract

With the development of community detection algorithm, traditional clustering methods are hardly capable enough to meet our needs. This paper proposes a new algorithm based on spectral method that can identify the overlapping and hierarchical communities in complex networks. Experiments on artificial and real datasets are conducted to verify the algorithm by comparing it with other algorithms. The experimental results show that our algorithm can efficiently identify overlapping communities in complex networks and find hierarchies of the communities.