A Distributed Framework for Community Analysis in Social Networks

A community in a network is a group of nodes which are densely connected internally and sparsely connected externally. In our study, we provide a novel, distributed approach for community detection. We propose a generic framework which describes behaviours of individual node in the network. Under this framework, the nodes are self-centred but produce a collective outcome to the network. Our approach may also be used to detect communities in dynamic networks. Our experiments include modularity comparison with an established algorithm, detecting communities on real networks and capturing changing communities on simulated dynamical networks