Community detection based on adaptive kernel affinity propagation

Detecting community structure in complex networks is a challenging problem which has attracted great interest in recent years. In this paper, a method called adaptive kernel affinity propagation is proposed to detect communities in networks, in which Markov diffusion kernel is transformed to implicitly measure the dissimilarities between different nodes and then adaptive affinity propagation is applied to determine the optimal number of communities and the corresponding membership assignment automatically. Experimental results on both computer-generated and real-world networks demonstrate that adaptive kernel affinity propagation can detect the correct and meaningful communities efficiently.