动态社会网络中的社区挖掘算法研究

Algorithm Research on Community Mining from Dynamic Social Networks

为解决社区挖掘问题,针对社会网络的动态特性,给出了新的社区定义,并结合连通性和频繁性概念提出一种新的算法DCSMA(Dynamic Community Structure Mining Algorithm)。挖掘时刻连通的个体集合作为社区,采用层状结构模型,根据重要性权重区分社区内个体,使社区结构更加清晰。在标准测试数据集上的实验结果表明了该算法的可行性和有效性。

Community mining is one of the hot research fields in social network analysis. In view of the dynamic character of social network, proposes a novel definition of community, and based on the concept of connectivity and frequency, it introduces a novel algorithm DCSMA （ Dynamic Community Structure Mining Algorithm）. It can successfully find the set of individuals which retain connectivity at any successive moment. In addition, we adopt the samdwich model, and distinguish the individuals according to their importance in order to make the structural framework of community more clear and better understood. Experiments on standard test data show that the algorithm is feasible and effective.