基于用户关联与主题关注的朋友圈兴趣组发现方法参考.pdfVIP

2016 年 第 26 卷 第 6 期 计 算 机 系 统 应 用 ① 基于用户关联与主题关注的朋友圈兴趣组发现方法 石小丹, 王海侠, 吴爱华 (上海海事大学 信息工程学院, 上海 201306) 摘 要: 传统社区发现算法大多考虑因素单一, 联系密切的友人间关注点可能差异较大, 而关注点相同的用户却 又可能不在一个朋友圈内. 为此, 提出了一种混合社区发现算法 HCDA, 它既考虑社区网络中的节点关注点, 又 考虑了社区网络拓扑结构, 以社区用户间的公共邻居比、关注度及发布微博相似度为依据, 度量相邻节点间的社 区关联紧密度. 并以此为基础, 依据相邻节点间的社区增益值, 迭代地扩展社区, 发现朋友圈中真正的兴趣小组. 实验表明, 相较于其他方法, 本算法能够更准确的发现社区. 关键词: 社区发现; 网络分裂; 社区网络拓扑结构; 微博主题模型; 朋友圈 Discovering Interest Groups among Friend Circle Based on Users’ Association and Common Topics SHI Xiao-Dan, WANG Hai-Xia, WU Ai-Hua (Information Engineering College, Shanghai Maritime University, Shanghai 201306, China) Abstract : Most traditional co unity detection algorithms always consider single factor. Friends who have close relationship may have different concerns and users who have co on concerns may not be in a circle of friends. To solve the problems, this thesis presents a hybrid co unity detection algorithm HCDA, which takes into account the concerns of the co unity network nodes, but also consider the topological structure of co unity network. On this basis, it expands iteratively the co unity by the co unity gain value between adjacent nodes to find the real interest groups among friend circles. The experimental results illustrate that compared with other methods the proposed algorithm can find the co unity more accurately. Key words: co unity detection; network splitting; co unity network topological structure; micro-blog topic model; circle of friends 微博, 作为一种