A Community Divisive Algorithm Using Local Weak Links

Detecting community structure can give a significant insight into structural and functional properties in complex networks. In this paper, we map nodes into random variables and use correlation coefficient to evaluate similarity between nodes. After defining the local weak links and analyzing the properties of correlation coefficient, we propose a divisive algorithm to uncover reasonable communities in the complex network. Simulation results in real and artificial networks show that the proposed algorithm is efficient in detecting community structures.