Abstract:  With the development of the research on complex networks, people have found many important properties about them. The community structure is one of the most popular and the most important topological properties of complex networks. So far, a large number of algorithms have been proposed to detect community structures in complex networks, but most of them are suitable for a specific network structure. Based on the previous research, this paper puts forward a kind of improved algorithm based on Affinity Propagation. First of all, dissimilarity distance matrix is calculated by using the shortest path algorithm. Then the exemplars are identified by the Affinity Propagation algorithm. Finally, based on the maximum modularity and minimum variation, “Community centers filtering” model is established to automatically identify the community structure of the network. Experimental results on real datasets, show that the proposed algorithm, comparing with the traditional method, has higher performance related to the accuracy, and the ability to tolerate the resolution limit.

Key words: shortest path, maximum modularity, community detection, affinity propagation