基于PageRank的微博用户影响力算法研究

随着Web技术的发展,微博已经成为最受欢迎的社交平台之一了。在中国,微博用户规模已经达到了2.42亿。微博用户影响力计算对社会信息在微博里面有效传播,正确传播,健康传播有着非常重要的意义。本文以新浪微博数据为实验的对象,通过改进传统的PageRank模型,提出了的新型用户影响力排名算法---MBUI-Rank(Micro-Blog User Influence Rank)算法。在考虑传统PageRank方法的用户链接关系的同时,MBUI-Rank算法还考虑到微博用户自身行为活动,构建用户对微博的影响的动态挖掘模型。实验结果表明,MBUI-Rank算法与传统的PageRank算法相比,可以更加真实有效地反映微博用户的实际影响力。

Research on the Algorithm of Micro-Blog User Influence Based on PageRank

With the development of Web technology, microblogging has become one of the most popular social platform. In China, the size of microblogging users has reached 242 million. Microblogging users influence the calculation of social information in the microblogging inside effective dissemination, correct communication, health communication has a very important significance. In this paper, a new user impact ranking algorithm named MBUI-Rank (Micro-Blog User Influence Rank) algorithm is proposed by improving the traditional PageRank model with the data of Sina Weibo as the experiment object. The MBUI-Rank algorithm takes into account the behavior of micro-blog users, and constructs a dynamic mining model of users" impact on micro-blogging while considering the user link relation of traditional PageRank method. The experimental results show that the MBUI-Rank algorithm can reflect the actual influence of micro-blog users more realistically and effectively than the traditional PageRank algorithm.