A community detection method based on ranking and heat diffusion algorithm in social networks

The past decade has witnessed a rapid development of social networks.Community detection is a very important problem in social network analysis. Finding a community in a social network is to identify a set of nodes such that they interact with each other more frequently than with the nodes outside the group. Classical clustering approach, K-means, has been shown to be very efficient to detection community. However it is sensitive to the initial seeds.To solve this problem, in this study, we first find K seeds using PageRank and second extract community structure of the network using heat diffusion similarity and K-means. Using heat diffusion similarity, we can get the network’s global information about any pair of vertices. The empirical study on real networks show that our algorithm in most social networks studied is better than algorithms including K-rank, K-means, BGLL, OSLOM ,LPA , Infomap and is efficient to be used for sparse large networks.