CUT: community update and tracking in dynamic social networks

Abstract

Social network exhibits a special property: community structure. The community detection on a social network is like clustering on a graph, but the nodes in social network has unique name and the edges has some special properties like friendship, common interest. There have been many clustering methods can be used to detect the community structure on a static network. But in real-world, the social networks are usually dynamic, and the community structures always change over time. We propose Community Update and Tracking algorithm, CUT, to efficiently update and track the community structure algorithm in dynamic social networks. When the social network has some variations in different timestamps, we track the seeds of community and update the community structure instead of recalculating all nodes and edges in the network. The seeds of community is the base of community, we find some nodes which connected together tightly, and these nodes probably become communities. Therefore, our approach can quickly and efficiently update the community structure.