Detecting community structure of complex networks based on network potential

Abstract

Community structure is an important topological property of complex networks, which is beneficial to understand the structures and functions of complex networks. In this chapter a new statistical parameter, which we call network potential, is proposed to measure a complex network by introducing the field theories of physics. We then present a detecting algorithm of community structure based on the network potential whose main strategy is partitioning the network by optimizing the value of the network potential. We test our algorithm on both computer-generated networks and real-world networks whose community structure is already known. The experimental results show the algorithm can be effectively utilized for detecting the community structure of complex network. © 2012 Springer Science+Business.