基于影响力与种子扩展的重叠社区发现

Overlapping Community Detection Based on Influence and Seeds Extension

社区发现作为复杂社交网络中一个重要的研究方向.针对目前基于种子节点的算法在种子选取与扩展等方面的不足,提出了一种基于影响力与种子扩展的重叠社区发现算法(Influence Seeds Extension Overlapping Community Detection,简称i-SEOCD算法).首先,利用节点影响力策略找出具有紧密结构的种子社区.其次,从这些种子社区出发,计算社区邻居集节点与社区的相似度,并取出相似度超过设定阈值的节点.然后,采用优化自适应函数的策略来扩展社区.最后,对网络中的自由节点进行社区隶属划分,进而实现了整个网络的重叠社区结构挖掘.在真实社交网络和人工生成网络上实验表明,i-SEOCD算法能够准确、快速地发现复杂网络中的重叠社区结构.

Community detection is a significant research direction in the research of social networks.To improve the quality of seeds selection and expansion,we propose an influence seeds extension overlapping community detection (i-SEOCD) algorithm for overlapping community detection.First,i-SEOCD uses a node influence strategy to find the seed communities with tight structures.Second,on the basis of the seed communities,we calculate the similarity among communities and their neighbor nodes.The nodes whose similarity is greater than a predefined threshold are selected.Third,the strategy of optimizing a self-adaptive function is adopted to expand the communities.Finally,the free nodes in the network are assigned to their corresponding communities in order to find out all the overlapping community structures.Experiments on the real and artificial networks show that i-SEOCD is capable of discovering overlapping communities in complex social networks efficiently.