用户生成内容中的垃圾意见研究综述*

随着互联网的快速发展，互联网用户更倾向于在博客、购物网站等场所借助于发表文章、评论等手段表达自己对感兴趣话题的关注、意见和相互交流。对这些用户生成内容进行挖掘可以得到很多宝贵的可应用于很多领域的知识，但其中存在的不请自来、与话题无关（广告等）或虽与话题相关但是虚假性的内容严重影响了挖掘的效果，并产生了误导用户、浪费资源等不良影响。因此，作为意见挖掘研究工作重要组成部分的垃圾意见研究工作开始得到空前的重视。从垃圾意见特征、垃圾意见检测方法两个方面对目前垃圾意见的研究状况进行了综述。

Survey on research of opinion spam in user generated content

With the rapid development of Internet,Internet users are more inclined to express their concerns,views for the interest topics and interact with each other in blog sites,ecommerce sites and other places by means of publishing articles,reviews and other means.And a lot of valuable knowledge mined from the user-generated-content can be applied to many fields.However,the contents which are uninvited,unrelated to the topics(e.g.advertising) or related to the topics but false in the user-generated-content have ...