Exploring and Detecting Opinion Spam on Social Media

In recent years, microblogging service such as Twitter and Weibo attracts a large number of users. Unlike traditional media by which user can only accept information, social media allows users to share their opinions. Among the social media users, there exist a group of users called opinion spam. They are well organized and post a large number of purposed comments to misdirect the public opinion. In this way, they significantly magnify the impact of their employers. We conduct quantitative analysis to study and understand the characteristics of opinion spam. We analyze the psycholinguistic styles of opinion spam, explore their behavior patterns and network structure. Finally, based on the analysis, context based collective classification is proposed to detect opinion spam and the model can achieve 91% F1 score.