Abstract
The value and credibility of online consumer reviews are compromised by significantly increasing yet difficult-to-identify fake reviews. Extant models for automated online fake review detection rely heavily on verbal behaviors of reviewers while largely ignoring their nonverbal behaviors. This research identifies a variety of nonverbal behavioral features of online reviewers and examines their relative importance for the detection of fake reviews in comparison to that of verbal behavioral features. The results of an empirical evaluation using real-world online reviews reveal that incorporating nonverbal features of reviewers can significantly improve the performance of online fake review detection models. Moreover, compared with verbal features, nonverbal features of reviewers are shown to be more important for fake review detection. Furthermore, model pruning based on a sensitivity analysis improves the parsimony of the developed fake review detection model without sacrificing its performance.
Acknowledgment
Any opinions, findings or recommendations expressed here are those of the authors and not necessarily those of the sponsor of this research.
Funding
This study was supported by the National Science Foundation (Award #695SES 1527684).
Notes
4. www.thestar.com/business/2013/09/23/fake_online_reviews_exposed_by_new_york_attorney_general.html.
5. www.nltk.org.
Additional information
Funding
Notes on contributors
Dongsong Zhang
DONGSONG ZHANG ([email protected]; corresponding author) received his Ph.D. in management information systems from the University of Arizona. He is a chair professor at the International Business School, Jinan University, China, and a full professor in the Department of Information Systems at the University of Maryland, Baltimore County. His research interests include social computing, mobile human–computer interaction, business analytics, health information technologies, and online communities. He has published approximately 140 papers in journals and conference proceedings, including Journal of Management Information Systems, MIS Quarterly, IEEE Transactions on Knowledge and Data Engineering, IEEE Transactions on Software Engineering, and others. His research has been funded by the U.S. National Science Foundation, the U.S. National Institutes of Health, the U.S. Department of Education, the National Natural Science Foundation of China, Chinese Academy of Sciences, and Google, Inc.
Lina Zhou
Lina Zhou ([email protected]) is an associate professor in the Department of Information Systems at the University of Maryland, Baltimore County. She received her Ph.D. in computer science from Peking University. Her research interests include online deception detection, computer-mediated communication, text mining, social network analysis, and intelligent user interfaces. She has published over sixty journal papers in Journal of Management Information Systems, MIS Quarterly, IEEE Transactions on Data and Knowledge Engineering, and others. Her research has been funded by the U.S. National Science Foundation.
Juan Luo Kehoe
Juan Luo Kehoe ([email protected]) is a master’s student in the Department of Information Systems at the University of Maryland, Baltimore County. She received her Ph.D. in biochemistry and molecular biology from China Agricultural University. Her research focuses on data science and big data analysis in various domains.
Isil Yakut Kilic
Isil Yakut Kilic ([email protected]) is a Ph.D. student in the Department of Information Systems at the University of Maryland, Baltimore County. She received her Master’s degree in computer engineering from Bilkent University, Turkey. Her research focuses on mining user-generated content and generalizing across domains.