基于rc 模型的微博社区用户影晌力分析 - 计算机科学.pdf

第 44 卷第3 期 计算机科学 Vo1. 44 NO.3 COMPUTER SCIENCE 岛1ar.2017 2017 年 3 月 基于 R-C 模型的微博社区用户影晌力分析 王振飞朱静阳郑志蕴宋玉 (郑州大学信息工程学院 郑州 450001) 摘 要 微博社区中用户的影响力对微博信息的有效传播具有重要意义。为了快速并准确地寻找微博社区信息传播 的规律，提出一种基于微博社区计算用户影响力的 USR 算法。首先提取种子用户的数据，利用 R-C 模型进行微博社 区发现，在划分好的社区中选取一个社区;然后依据 USR 算法，对社区内的用户进行影响力计算;最后输出用户的影 响力。以新浪微博数据集为例，提出孤立点的概念和信息传播实际影响人次覆盖率评价指标，将 USR 算法与传统影 响力算法进行对比。实验结果表明，使用 USR 算法能够得到较优的结果。 关键词微博，影响力，R-C 模型，USR 算法，信息传播 中图法分类号 TP311 文献标识码 A 001 10. 11896/j. issn. 1002-137叉 2017.03.052 Analysis of Microblog Community Users Influence ßased on R-C Model WANG Zhen-fei ZHU Jing-yang ZHENG Zhi-归n SONG Yu (School of Information Engineering ,Zhengzhou University ,Zhengzhou 450001 ,China) Abstract The microblog community users influence has great significance in effective dissemination of microblog in formation. To rapidly and accurately find the regularity of micro-blog community s information dissemination ,a micro- blog community users importance algorithm was presented. Firstly ,seed user data are extracted and microblog commu- nities are detected by using R-C model ,and one community form the divided communities is selected. Then ,the influence of user in the community is calculated according to USR algorithrn. Finally , the influence of user in the community is outputted. Taking the sina microblog data sets as example ,the concept of isolated point and the coverage evaluation in dex of information dissemination impact person-time were proposed. We computed the users influence by comparing USR algorithm with other traditional algorithms. Experiments show that the USR algorithm can