计算机科学与探索 ›› 2013, Vol. 7 ›› Issue (9): 838-847.DOI: 10.3778/j.issn.1673-9418.1305009

• 学术研究 • 上一篇    下一篇

博客作者声誉度分析

杨  亮,许  侃,林鸿飞+,王  健,刁宇峰   

  1. 大连理工大学 信息检索研究室,辽宁 大连 116024
  • 出版日期:2013-09-01 发布日期:2013-09-04

Analysis of Blog Author Reputation

YANG Liang, XU Kan, LIN Hongfei+, WANG Jian, DIAO Yufeng   

  1. Information Retrieval Laboratory, Dalian University of Technology, Dalian, Liaoning 116024, China
  • Online:2013-09-01 Published:2013-09-04

摘要: 博客作为一种新兴的网络媒体,它具有自媒体的特征,使其成为舆情产生和传播的主要场所,因此对于博客作者和评论的研究是十分重要的。传统的博客作者声誉度排名是基于点击率和评论数量进行的,在此基础上结合评论的质量和博文的内容实现了对博客作者声誉度的排名。首先通过情感语义特征分析博文,获得博客的情感基调;然后利用基于段落的多句联合评估方法进行评价;最后结合得到的评论倾向性和质量及浏览量等信息获得作者的声誉度。实验结果表明,该方法可以更为有效地对博客作者的声誉度进行排名。

关键词: 评论质量, 情感倾向性, 多句联合评估, 博客作者声誉度

Abstract: Blog is a new kind of self-media, and becomes the main place of public opinion production and propagation. As a result, it is very important to do some research on Blog posts and reviews. Some ranking lists for blog authors are just calculated on click-through rate and the number of reviews. This paper focuses on combining the content of blog post and reviews into the click-through rate and the number of reviews to analyze the author reputation. Firstly, it gets the emotional tone of blog post by emotional feature analysis. Next, it evaluates the emotional type of blog post by sentences joint assessment method. Finally, it gives the author reputation degree based on the tendency of reviews, the quality of contents and the number of clicks. The experimental results show that it is more effective than the click rank of authors and makes more useful information about authors.

Key words: review quality, emotional tendency, sentences joint assessment method, blog author reputation