点击次数：

论文类型：期刊论文

发表时间：2013-01-01

发表刊物：Journal of Computational Information Systems

收录刊物：Scopus

卷号：9

期号：24

页面范围：9715-9724

ISSN号：15539105

摘要：Community detection is always an outstanding problem in the study of complex networks. We propose a novel hierarchical algorithm based on topological potential in the data field to detect community structures. At the beginning, being regarded as influence, the topical potential of each node is calculated using the networks topological structure and node information. Then, based on swarm intelligence and the mechanism of Priority to the Weak, we introduce a new clustering method to divide nodes into different groups. For each node, two kinds of effects, cognitive component and social component, are taken into account. Furthermore, the same partition method is employed to each community obtained until all nodes belong to the same community. The definitive features in our algorithm are that no predefined parameter is needed and no optimization objective function such as modularity is employed. The application in computer-generated networks and real-world networks demonstrates that the algorithm achieves good performance. Copyright ? 2013 Binary Information Press.