基于行为和社团的微博用户传播影响力分析  

Analyzing micro-blog users' propagation influence based on behavior and community

作　　者：陈振春 刘学军 李斌 Chen Zhenchun;Liu Xuejun;Li Bin(School of Computer Science ＆ Engineering,Nanjing Tech University,Nanjing 211816,China)

机构地区：[1]南京工业大学计算机科学与技术学院,南京211816

出　　处：《计算机应用研究》2018年第7期2075-2078,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(61203072);江苏省重点研发计划资助项目(BE2015697)

摘　　要：为快速、准确地识别微博网络中具有较大影响力的节点,提出了一种基于用户关系、行为以及社团结构的影响力评价算法。根据模块度对加权有向微博网络进行社团划分,综合微博网络中用户的粉丝数量、粉丝质量以及跨社团数目等特性度量节点的影响力。同时,对粉丝质量进行深入的分析,利用粉丝对用户的关注度作为粉丝质量的分配标准,最终完成用户传播影响力的评价。实验结果表明,该算法显著提高了评估用户传播影响力的准确性。To identify the nodes with great influence in the micro-blog network quickly and accurately,this paper proposed an influence evaluation algorithm based on user relation,behavior and community structure. First,according to the modularity,it divided the weighted directed micro-blog network into several communities. Second,it utilized the features of user in microblog network to measure the impact of nodes. These features included the number of fans,the quality of fans and the number of cross-community. Meanwhile a deep research on the quality of fans,the attention of fans to the user is the distribution standards of fans＇ quality. Finally,this paper completed the evaluation about user propagation influence. The experimental results show that the algorithm improves the accuracy of evaluating users＇ propagation influence obviously.

关 键 词：关系 行为 社团结构 影响力 微博网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]