计算机工程与应用 ›› 2016, Vol. 52 ›› Issue (16): 95-99.

• 大数据与云计算 • 上一篇    下一篇

基于SRank的社交网络影响力分析

任留名,李  廉,唐敏龙   

  1. 合肥工业大学 计算机与信息学院,合肥 230009
  • 出版日期:2016-08-15 发布日期:2016-08-12

Analysis of social networks influence based on SRank

REN Liuming, LI Lian, TANG Minlong   

  1. College of Computer and Information, Hefei University of Technology, Hefei 230009, China
  • Online:2016-08-15 Published:2016-08-12

摘要: 针对社交网络中用户影响力的评价问题,提出了一种基于SRank的评价算法。基于从社交网络中收集的大规模数据集,结合最近社会学理论研究成果分析PageRank及其改进算法应用于此场景中的不足。在此基础上总结社交网络中信息传播的规律,将用户与社交网络的关系强度定义为用户的人缘值,用来表示用户作为粉丝的信息再传播能力。然后提出了一个通过预测用户信息传播能力大小来分析和度量用户影响力的SRank用户影响力模型。在同样的数据集下相对于PageRank及其改进算法,SRank用户影响力模型获得了更好的影响力预测结果。基于大规模数据的实验结果表明,提出的方法是较为有效的。

关键词: 用户影响力, PageRank, 社会计算, 幂律分布, 150定律

Abstract: An evaluation algorithm based on SRank is proposed to evaluate the users’ influence in social networks. After collecting large?scale?data?sets and recent theoretical?research?results of sociology, there lie some disadvantages in the application?of?PageRank?user?influence?model. Besides, the rules of information?in social?networks is analyzed, it is defined that?the?strength of relationship?between?the?user?and?the social?network as the Value of User Popularity(VUP) which is used to describe user’s information re communication ability. It is put forward a user influence model based on SRank algorithm which can measure user influence depending on predicting user?information communication?ability. Compared with PageRank?user?influence?model, the SRank algorithm user influence model gets better predicting results on user influence. So the new model is effective.

Key words: user influence, PageRank, social computing, power-law distribution, rule of 150