文献类型：期刊文献

中文题名：基于贝叶斯后验模型的局部社团发现

英文题名：Local Community Detection Based on Bayesian Posterior Model

作者：王伟[1];程华[1];房一泉[1]

机构：[1]华东理工大学信息科学与工程学院,上海200237

年份：2014

卷号：40

期号：5

起止页码：619

中文期刊名：华东理工大学学报（自然科学版）

外文期刊名：Journal of East China University of Science and Technology

收录：CSTPCD;;Scopus;北大核心:【北大核心2011】；CSCD:【CSCD2013_2014】；

语种：中文

中文关键词：社团发现;贝叶斯后验模型;局部社团;BS模块度

外文关键词：community detection; Bayesian posterior model; local community； BS modularity

摘要：基于节点的局部社团发现在大数据社会网络分析中非常重要。针对Newman模块度在社团发现中的局限性,基于贝叶斯后验模型提出了BS模块度度量法。该方法结合节点的模块度和推荐概率进行建模,并以邻接并入为框架得到了一种新的局部社团发现算法。该方法克服了Newman模块度在稀疏网络中区分度低的问题以及社团结构差异大的分辨率问题,有效地寻找大规模网络中的局部社团。通过与Newman模块度在真实社团中的比较,验证了该度量方法的有效性。
The node-centric local community detection plays an important role in the analyis of big data social network. Aiming at the shortcoming of community detection with Newman modularity, an BS modularity based on Bayesian posterior model is proposed in this work to obtain a new local community detection method. It combines Newman modularity with nodes＇ recommending probabilities and takes adjacency merge as the framework. It is shown that the proposed algorithm can overcome the shortcomings of the Newman modularity, e. g., lower differentiation in sparse network and worse resolution on community structure and obtain the local community in large scale network. Comparing experiments with Newman modularity on benchmark data validate the BS modularity.