Community Evolution Mining in Dynamic Social Networks

Abstract Data that encompasses relationships is represented by a graph of interconnected nodes. Social network analysis is the study of such graphs which examines questions related to structures and patterns that can lead to the understanding of the data and predicting the trends of social networks. Static analysis, where the time of interaction is not considered (i.e., the network is frozen in time), misses the opportunity to capture the evolutionary patterns in dynamic networks. Specifically, detecting the community evolutions, the community structures that changes in time, provides insight into the underlying behaviour of the network. Recently, a number of researchers have started focusing on identifying critical events that characterize the evolution of communities in dynamic scenarios. In this paper, we present a framework for modeling and detecting community evolution in social networks, where a series of significant events is defined for each community. A community matching algorithm is also proposed to effciently identify and track similar communities over time. We also define the concept of meta community which is a series of similar communities captured in different timeframes and detected by our matching algorithm. We illustrate the capabilities and potential of our framework by applying it to two real datasets. Furthermore, the events detected by the framework is supplemented by extraction and investigation of the topics discovered for each community.