Paper: TopicSpam: a Topic-Model based approach for spam detection

ACL ID P13-2039
Title TopicSpam: a Topic-Model based approach for spam detection
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

Product reviews are now widely used by individuals and organizations for decision making (Litvin et al., 2008; Jansen, 2010). And because of the profits at stake, peo- ple have been known to try to game the system by writing fake reviews to promote target products. As a result, the task of de- ceptive review detection has been gaining increasing attention. In this paper, we pro- pose a generative LDA-based topic mod- eling approach for fake review detection. Our model can aptly detect the subtle dif- ferences between deceptive reviews and truthful ones and achieves about 95% ac- curacy on review spam datasets, outper- forming existing baselines by a large mar- gin.