论文部分内容阅读
Community Detection in social networks is usually considered as an objective optimization problem.Limited to the objective function,the global optimum cannot describe the real partition well,and it is time consuming.In this paper,a layered optimization framework is designed to improve the optimization process,reduce the scale of network and increase the quality of solution.The framework consists of three parts: finding cores in networks,repairing isolated nodes and optimization in a new constructed weighted network which is a compressed network of the origin one.Firstly,the equivalency of modularity optimization in the new compressed weighted network and the original one is proved.Furthermore,a combined algorithm of community Detection named DBPSO including similarity-based clustering,isolated nodes repairing strategies and a modified particle swarm optimization is proposed according to the layered optimization framework.In addition,a suitable mutation strategy for particle swarm optimization (PSO) is introduced to guarantee the convergence and global search ability.Finally,the experiments are conducted to evaluate the proposed algorithm by using the synthetic and real-world network datasets.The results show that the proposed algorithm can effectively extract the intrinsic community structure of social networks.