The community detection algorithm based on transferring the similarity

Abstract

For global community detection algorithm having high computation complexity and local community detection algorithm working bad in community quality, in this paper we propose an algorithm by finding the core nodes in different communities first, then use an algorithm based on transferring the similarity we propose in this paper to measure the similarity between core nodes with other nodes. Finally we divide the network by the similarity calculation results.The proposed algorithm is tested on both our network and common networks, and is compared with the typical algorithms in community detection. Experimental results verify and confirm the feasibility and validity of the proposed algorithm.