The Enemy in Your Own Camp: How Well Can We Detect Statistically-Generated Fake Reviews – An Adversarial Study

Online reviews are a growing market, but it is struggling with fake reviews. They undermine both the value of reviews to the user, and their trust in the review sites. However, fake positive reviews can boost a business, and so a small industry producing fake reviews has developed. The two sides are facing an arms race that involves more and more natural language processing (NLP). So far, NLP has been used mostly for detection, and works well on human-generated reviews. But what happens if NLP techniques are used to generate fake reviews as well? We investigate the question in an adversarial setup, by assessing the detectability of different fake-review generation strategies. We use generative models to produce reviews based on meta-information, and evaluate their effectiveness against deceptiondetection models and human judges. We find that meta-information helps detection, but that NLP-generated reviews conditioned on such information are also much harder to detect than conventional ones.

[1]  Jure Leskovec,et al.  No country for old members: user lifecycle and linguistic change in online communities , 2013, WWW.

[2]  F ChenStanley,et al.  An Empirical Study of Smoothing Techniques for Language Modeling , 1996, ACL.

[3]  Noah A. Smith.  Adversarial Evaluation for Models of Natural Language , 2012, ArXiv.

[4]  Claire Cardie,et al.  Finding Deceptive Opinion Spam by Any Stretch of the Imagination , 2011, ACL.

[5]  Massimo Poesio,et al.  Identifying fake Amazon reviews as learning from crowds , 2014, EACL.

[6]  David Yarowsky,et al.  Exploring Demographic Language Variations to Improve Multilingual Sentiment Analysis in Social Media , 2013, EMNLP.

[7]  Daniel Jurafsky,et al.  Generating Recommendation Dialogs by Extracting Information from User Reviews , 2013, ACL.

[8]  Bing Liu,et al.  Review spam detection , 2007, WWW '07.

[9]  Jure Leskovec,et al.  Learning Attitudes and Attributes from Multi-aspect Reviews , 2012, 2012 IEEE 12th International Conference on Data Mining.

[10]  Ee-Peng Lim,et al.  Finding unusual review patterns using unexpected rules , 2010, CIKM.

[11]  Dirk Hovy,et al.  User Review Sites as a Resource for Large-Scale Sociolinguistic Studies , 2015, WWW.

[12]  Dirk Hovy,et al.  Demographic Factors Improve Classification Performance , 2015, ACL.

[13]  Georgios Zervas,et al.  Fake It Till You Make It: Reputation, Competition, and Yelp Review Fraud , 2015, Manag. Sci..

[14]  Theodoros Lappas,et al.  Fake Reviews: The Malicious Perspective , 2012, NLDB.

[15]  Jo Mackiewicz,et al.  Reviewer Motivations, Bias, and Credibility in Online Reviews , 2008, Handbook of Research on Computer Mediated Communication.

[16]  Noah A. Smith,et al.  Narrative framing of consumer sentiment in online restaurant reviews , 2014, First Monday.

[17]  Ee-Peng Lim,et al.  Detecting product review spammers using rating behaviors , 2010, CIKM.

[18]  Dirk Hovy,et al.  Learning Whom to Trust with MACE , 2013, NAACL.

[19]  Sameer Badaskar,et al.  Identifying Real or Fake Articles: Towards better Language Modeling , 2008, IJCNLP.

[20]  M. Kenward,et al.  An Introduction to the Bootstrap , 2007 .