Community Detection by Affinity Propagation with Various Similarity Measures

Modularity or community structure is a natural characteristic in many real networks. The detection of community structure in complex networks can enhance the insight into the intrinsical structure of networks and then has become a key problem in the study of networked systems. we propose a method based on affinity propagation (AP) clustering for detecting communities in complex networks. We first evaluate several similarity measures, such as diffusion kernel similarity, shortest path based similarity on several widely well studied networks. Then, we apply AP with diffusion kernel similarity to three large biological networks, which demonstrates that our method can find biologically meaningful functional modules.