《计算机应用研究》|Application Research of Computers

基于用户特征属性的微博话题关键用户挖掘

Key users mining in micro-blogging topic based on user attributes

免费全文下载 (已被下载 次)  
获取PDF全文
作者 柯阳,隋杰
机构 中国科学院大学 工程科学学院,北京 100049
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)06-003-1614-04
DOI 10.19734/j.issn.1001-3695.2017.12.0795
摘要 针对微博话题存在时效性的特征以及用户之间交互行为特征,在经典PageRank算法的基础上,提出了基于用户交互的微博用户挖掘算法来有效挖掘推动微博话题流行的关键用户。首先介绍了微博话题关键用户的定义及其相关特征;其次,由于传统模型未考虑用户交互以及时间属性的影响,所以融合了时间属性以及用户之间交互特征,同时结合微博网络结构提出了MUR算法;最后,将算法与经典PageRank算法和TS算法作了比较。实验结果表明,模型在反映微博话题用户的时效性、话题推动以及对粉丝的影响力等方面表现较好,证明了模型的合理性和有效性。
关键词 关键用户; 微博用户排序; 时间属性; 用户交互
基金项目 国家重点研发计划项目(2017YFB0803001)
国家自然科学基金面上项目(61572459)
本文URL http://www.arocmag.com/article/01-2019-06-003.html
英文标题 Key users mining in micro-blogging topic based on user attributes
作者英文名 Ke Yang, Sui Jie
机构英文名 School of Engineering Science,University of Chinese Academy of Sciences,Beijing 100049,China
英文摘要 Considering the timeliness of the microblogging topic and the feature of interaction between the users, on the basis of classical PageRank algorithm, this paper put forward a key user's mining algorithm based on user interaction to effectively find topic-sensitive key users. Firstly, this paper introduced the definition of key users in microblog topic and its relevant characteristics. Secondly, in that the traditional models ignored the influence of user interaction and time attribute, this model fused the time property and the characteristics of interaction between the user together in this model at the first time and then it put forward the MUR algorithm with the combination of the microblogging network structure. Finally, it compared the algorithm with the classical PageRank algorithm and TS algorithm. The experimental results show that the model is more reasonable in terms of timeliness and topic driving, certificating the rationality and validity of the model.
英文关键词 key user; MUR(microblog user rank); time property; user interaction
参考文献 查看稿件参考文献
 
收稿日期 2017/12/5
修回日期 2018/1/25
页码 1614-1617,1622
中图分类号 TP391
文献标志码 A