TY - JOUR
T1 - Creating and detecting fake reviews of online products
AU - Salminen, Joni
AU - Kandpal, Chandrashekhar
AU - Kamel, Ahmed Mohamed
AU - Jung, Soon gyo
AU - Jansen, Bernard J.
N1 - Publisher Copyright:
© 2021 The Authors
PY - 2022/1
Y1 - 2022/1
N2 - Customers increasingly rely on reviews for product information. However, the usefulness of online reviews is impeded by fake reviews that give an untruthful picture of product quality. Therefore, detection of fake reviews is needed. Unfortunately, so far, automatic detection has only had partial success in this challenging task. In this research, we address the creation and detection of fake reviews. First, we experiment with two language models, ULMFiT and GPT-2, to generate fake product reviews based on an Amazon e-commerce dataset. Using the better model, GPT-2, we create a dataset for a classification task of fake review detection. We show that a machine classifier can accomplish this goal near-perfectly, whereas human raters exhibit significantly lower accuracy and agreement than the tested algorithms. The model was also effective on detected human generated fake reviews. The results imply that, while fake review detection is challenging for humans, “machines can fight machines” in the task of detecting fake reviews. Our findings have implications for consumer protection, defense of firms from unfair competition, and responsibility of review platforms.
AB - Customers increasingly rely on reviews for product information. However, the usefulness of online reviews is impeded by fake reviews that give an untruthful picture of product quality. Therefore, detection of fake reviews is needed. Unfortunately, so far, automatic detection has only had partial success in this challenging task. In this research, we address the creation and detection of fake reviews. First, we experiment with two language models, ULMFiT and GPT-2, to generate fake product reviews based on an Amazon e-commerce dataset. Using the better model, GPT-2, we create a dataset for a classification task of fake review detection. We show that a machine classifier can accomplish this goal near-perfectly, whereas human raters exhibit significantly lower accuracy and agreement than the tested algorithms. The model was also effective on detected human generated fake reviews. The results imply that, while fake review detection is challenging for humans, “machines can fight machines” in the task of detecting fake reviews. Our findings have implications for consumer protection, defense of firms from unfair competition, and responsibility of review platforms.
KW - Detection
KW - Fake reviews
KW - Marketing
KW - e-commerce
KW - eWOM
UR - http://www.scopus.com/inward/record.url?scp=85115082993&partnerID=8YFLogxK
U2 - 10.1016/j.jretconser.2021.102771
DO - 10.1016/j.jretconser.2021.102771
M3 - Article
AN - SCOPUS:85115082993
SN - 0969-6989
VL - 64
JO - Journal of Retailing and Consumer Services
JF - Journal of Retailing and Consumer Services
M1 - 102771
ER -