Modularity Modeling and Evaluation in Community Detecting of Complex Network Based on Information Entropy

Community structure is the most basic and important topologic characteristic of complex network and community detection method is therefore significant to its character statistics. A new theoretic model of modularity Q based on information entropy (IE) with low complexity and better accuracy is proposed to promote clustering accuracy. IE algorithm reaches better community detecting results than GN and FastGN algorithm on network with definite community structure and unidentified community structure, while computation complexity declines. The simulation results mentioned above show that IE will find more accurate community structure than traditional methods of edge rate on most classic complex network dataset. According to simulation result compared with the seven main classic community detecting methods on simulated random networks and real networks, information entropy based modularity is more accurate than traditional modularity of edge rate.