A Blog Advertising Comments Detecting Approach Based on Behavior Analysis

摘要：

At present, there are many annoying advertisements in blog comments. In this paper, we present an approach to detect the blog advertising comments. Through investigating the blog advertising comments, we found that the behavior features of fake writers (writers of advertisement comments) have obviously different from the real writers (writers of ham comments). The writing interval, writing style and writing content are used to measure the differences of the behavior features between the fake and real writers. The experimental results reveal the proposed method can effectively and flexibly detect the advertising comments in blog.

关键词： blog advertising comments behavior analysis

作者: Minglong Yi Ming Xu

作者单位: Institute of Computer Application Technology Hangzhou Dianzi University Hangzhou, China

会议类型: 国际会议

会议名称: 2010 Second Asia-Pacific Conference on Information Processing(2010年第二届亚太地区信息处理国际会议 APCIP 2010)

会议地点: 南昌

会议语种:英文

页码: 134-137

在线出版日期: 2010-09-17（万方平台首次上网日期，不代表论文的发表时间）