基于标签传播的社区挖掘算法研究综述

摘要

Social networks have been a hot area of research because of its popularity. Discover potential communities in social networks through community mining,and find community structures that have more realistic significance through detecting overlapping communi-ties. However there is lot of inconvenience because of the sheer amount of data in social networks. So fast algorithm for mining communi-ty are getting more and more attention. The algorithms based on the thoughts of label propagation have nearly linear time complexity. In this paper,study the algorithms based on the thoughts of label propagation from various aspects and analyze those algorithms' improve-ment ideas in the future research.%社会网络由于其流行程度已经成为众多学者的研究热点。通过社区挖掘算法可以发现存在于社会网络中的潜在社区，而重叠社区挖掘则可以挖掘出更具有现实意义的社区结构。但是在研究中社会网络所包含的庞大数据量又会为之带来种种不便，因此快速的社区挖掘算法就受到了越来越多的重视。基于标签传播的社区挖掘算法具有近乎线性的时间复杂度。文中将从多方面研究目前基于标签传播的社区挖掘算法的优劣，并且详细分析基于标签传播算法在以后研究中的改进思路。