Computer Engineering and Applications ›› 2015, Vol. 51 ›› Issue (4): 128-132.

Previous Articles     Next Articles

Blogger influence ranking algorithm based on user quality in Sina microblog community

TANG Feilong1, YE Shiren2, XIAO Chun1   

  1. 1.Key Lab of Intelligent Computing & Infor Processing of MOE, Xiangtan University, Xiangtan, Hunan 411105, China
    2.School of Information Science & Engineering, Changzhou University, Changzhou, Jiangsu 213164, China
  • Online:2015-02-15 Published:2015-02-04

基于用户质量的微博社区博主影响力排序算法

唐飞龙1,叶施仁2,肖  春1   

  1. 1.湘潭大学 信息工程学院智能计算与信息处理教育部重点实验室,湖南 湘潭 411105
    2.常州大学 信息科学与工程学院,江苏 常州 213164

Abstract: With the extraordinary spreading speed and impacting scope, has microblog become most important social network. At present, Sina Weibo(Chinese microblog) has 300 million plus users and is listed in the top rank of all social sites, which exhibits distinguished interaction and user behaviors. The primary investigation is performed among Chinese microblog using Sina Weibo. The user network graph is utilized to present the Chinese micro-blog network society, by analyzing its characteristics as a new social network. A User Impact Rank(UIR) measure is developed based on PageRank. Bloggers’ UIR is dynamically refined through the user relative micro-force value and the relative link quality. A test corpus is built by collecting all blogers and their posts in an active microblog community to evaluat the UIR measure method. The experimental results show that the proposed UIR can effectively exclude the illusive influence from spurious followers.

Key words: relative micro-force value, relative link quality, PageRank, User Impack Rank

摘要: 微博特有的移动终端轻博客发布与交互模式,使其迅速成为使用范围最大、影响力最大的社交媒体。新浪中文微博现有超过3亿用户,发展最为迅速,中文微博和其他微博相比具有独特性,一些大“V”博主的影响力堪比电台电视。通过分析微博的网络结构特征,总结出微博相对于其他传统社会载体的特性。利用PageRank算法的思想,设计了基于用户质量的User Impack Rank(UIR)排序算法。UIR算法通过用户相对微力值和用户相对链接质量对各博主的影响力进行动态的评估。在一个活跃的微博社区数据集上进行了全面的实验,实验结果显示了UIR算法能更加准确和客观地对用户的影响力进行排序,并且能有效地消除僵尸粉丝对排序的影响。

关键词: 相对微力值, 相对链接质量, PageRank, User Impack Rank