Classification of Artificial Intelligence-Generated Product Reviews on Amazon
Jia-Luen Yang · 2025
Amazon has been flooded with artificial intelligence (AI)-generated product reviews that offer minimal value to customers. These AI reviews merely echo the given product descriptions without providing any authentic information on how buyers feel when using the products. Therefore, an AI review-identifying method was developed to enhance the quality of the review-reading experience in this study. A dataset of 6217 Amazon reviews was compiled including 1116 identified as AI-generated ones. They were classified with a 99.25% F1 score on the test data using the term frequency–inverse document frequency (TF–IDF) and support vector classifier (SVC). The developed method enables the detection of AI-generated reviews on the internet, fostering an authentic and reliable platform.