Title: Enhanced accuracy of detecting fraudulent product reviews using a fusion machine learning approach

Authors: Pallavi Zambare; Ying Liu

Addresses: Texas Tech University, Lubbock, USA ' Texas Tech University, Lubbock, USA

Abstract: Spam product reviews are fake or misleading reviews posted to promote a product or service, or to damage a competitor's reputation. They can be difficult to detect, as they are often written to appear legitimate and may include fake profiles or misleading information. Spam product reviews can be harmful to businesses and consumers. Fake reviews and ratings can damage credibility and trustworthiness by misleading consumers about the quality of a product. They can also harm their trust in the review process. It is important for businesses and review platforms to have measures in place to identify and eliminate spam product reviews. Here, we proposed a hybrid technique for identifying fake spam product reviews. This paper first introduces the task of spam online product reviews detection and makes a common definition of spam reviews. Then, we comprehensively conclude the existing method on publicly available datasets. Finally, we have shown the performance comparison for traditional machine learning, deep learning and proposed hybrid classifier. Based on the evaluation it shows that hybrid of CNN and random forest outperforms others.

Keywords: fake reviews detection; methodologies; machine learning; web-scrapping; information retrieval.

DOI: 10.1504/IJSEM.2025.148473

International Journal of Services, Economics and Management, 2025 Vol.16 No.4/5, pp.380 - 406

Received: 24 Mar 2023
Accepted: 05 Jul 2023
Published online: 08 Sep 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article