An Improved Community Detection Algorithm Based on Local Information in Social Networks

Abstract

As community detection having been a hot issue in recent years, firstly the correlate clustering technologies based on local information are studied and summarized. Secondly, the label propagation algorithm that is short for LPA is researched and analyzed in depth. Finally, for the random strategy of LPA results that the network partition is always not optimal, we introduces the concept of the similarity of nodes' attributes, and proposes label propagation algorithm on the basis of the similarity of nodes' attributes called LPA-SNA for short.