Real Time Distributed Community Structure Detection in Dynamic Networks

Abstract

Communities can be observed in many real-world graphs. In general, a community can be thought of as a portion of a graph in which intra-community links are dense while inter-community links are sparse. Automatic community structure detection has been well studied in static graphs. However, many practical applications of community structure involve networks in which communities change dynamically over time. Several methods of detecting the community structure of dynamic graphs have been proposed, however most treat the dynamic graph as a series of static snapshots, which creates unrealistic assumptions. Others require large amounts of computational resources or require knowledge of the dynamic graph from start to finish, relegating them to post-processing. For those who desire real-time community structure detection distributed over the observing network, these solutions are insufficient. This paper proposes a new method of community structure detection which allows for real time distributed detection of community structure.