情报杂志2016，Vol.35Issue(11)：183-188,6.DOI:10.3969/j.issn.1002-1965.2016.11.033
基于交互行为的在线社会网络社区发现方法研究
New Method of Community Detection for Online Social Networks Based on Interactive Behaviors
摘要
Abstract
Purpose/Significance] Community detection is an important and challenging problem in several research fields such as social complex networks, biology gene expression, and physics systems and so on. There are two popular types of community detection for the online social network in general nowadays. The first one is based on the statistic relationships between nodes. The second one is based on the dynamical relationships between nodes. The dynamical interactive relationship can better reflect the real relationship between social media users. [ Method/Process] A new method of community detection for online social networks is proposed based on the interactive be-havior of the nodes of networks in this paper. Based on the four aspects of the interaction between users such as sharing, commenting, a-greeing and mentioning relationship, a weighted interactive network is constructed. Those nodes that have the maximum and positive mod-ule gain are merged until all the module gains are negative. [ Result/Conclusion] Compared to exiting methods, the method has two merits such as the module function is improved and interactions of nodes are measured by formula. At last, experiments on the data from Sina Weibo verify the proposed method and show its effectiveness.关键词
在线社交网络/社区发现/交互行为/模块度函数/动态网络/模块度增益Key words
online social network/community detection/interactive behavior/module function/dynamic network/module gain分类
信息技术与安全科学引用本文复制引用
田博,凡玲玲..基于交互行为的在线社会网络社区发现方法研究[J].情报杂志,2016,35(11):183-188,6.基金项目
国家自然科学基金项目“BtoC电子商务中基于分层Bayesian网络的信任与声誉计算理论研究”(编号:71302080) （编号:71302080）
教育部人文社科基金“移动支付中的感知信任与消费决策理论研究”(编号:13YJC630149)资助。 （编号:13YJC630149）
