摘要
博客的出现丰富和改变了网络的内涵,影响了人们的信息传递方式,同时博客评论作为一种交互方式在博客中广泛存在,给信息监管带来了新的问题。通过分析现有的博客过滤系统,将广泛应用于文本过滤的贝叶斯方法应用到博客评论中,针对博客评论中广泛存在的广告机器人特点,结合信息指纹对其进行识别和过滤。同时对影响博客评论过滤效果和执行速度的指纹函数进行了分析讨论和实验对比,实验结果表明基于贝叶斯方法和信息指纹相结合的博客评论过滤是行之有效的,而且相对于单独的贝叶斯方法更有利于提高系统运行效率和发现广告机器人现象。
The appearance of blog enriches and changes the network＇s connotation, and influences the ways of informafion-delivering.Blog criticism,as an exchanging way,has been widely used in blog and thus brings new problems to information warding. This paper on one hand, applies Bayes of text filtering in blog criticism by analysis of blog filtering system in hand;On the other hand,because of the specific features of robot widely existing in blog criticism,this paper recognizes and fdters the criticism combining the information fingerprint.Moreover,this paper analyzes and discusses the fingerprint functions that influence blog-filtering＇s effect and carrying-out speed.The result of this experiment shows that this blog-filtering is effective, based on Bayes and informafion fingerprint,and is more advanced than the only Bayes in improving system running efficiency and finding out the phenomenon of advertisement robot.
出处
《计算机工程与应用》
CSCD
北大核心
2008年第24期159-161,180,共4页
Computer Engineering and Applications
关键词
博客
贝叶斯
评论
信息指纹
blog
Bayes
comments
information fingerprint
