International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064


Downloads: 1 | Views: 133 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1

Research Paper | Computer Science | India | Volume 11 Issue 12, December 2022 | Rating: 5 / 10


Detecting Fraudulent Reviewers on eCommerce Platforms

Devanathan Iyer | Ratna Krishnaswamy | Suresh Tripathy | Anant Bhanushali


Abstract: Fraudulent online reviews have become a growing concern for eCommerce platforms and their users. These fake reviews, often left by incentivized reviewers, can mislead consumers and harm the credibility of the platform. In this paper, we propose a machine learning-based approach to detecting fraudulent reviewers on eCommerce platforms. Our approach utilizes various features derived from the reviewers? profile, reviewing activity and other behaviour to train a binary classifier. These features include aspects such as the writing style and sentiment, the frequency and timing of reviews, and the diversity of products reviewed. We evaluate our method on a dataset of reviews from a major eCommerce platform and compare its performance with traditional techniques such as rule-based methods and simple statistical models. Our results show that our machine learning-based approach outperforms traditional techniques in detecting fraudulent reviewers. The classifier achieves an F1 score of 0.87 on the test set, demonstrating high precision and recall. Additionally, our approach is able to adapt to changing patterns of fraud, making it more robust against evolving fraudster tactics. Our study provides evidence that machine learning-based approaches can be effective in detecting fraudulent reviewers on eCommerce platforms. Our approach offers a promising solution to the problem of fake reviews and can be integrated into the review moderation process to improve the accuracy and efficiency of fraud detection. Future work can extend our approach to incorporate additional features and to consider more complex forms of review fraud.


Keywords: eCommerce, fake reviews, fraud, incentivized review


Edition: Volume 11 Issue 12, December 2022,


Pages: 1275 - 1279


How to Download this Article?

Type Your Valid Email Address below to Receive the Article PDF Link


Verification Code will appear in 2 Seconds ... Wait

Top