Key points are not available for this paper at this time.
The proliferation of online reviews has made them a vital source of information for consumers in making purchasing decisions. However, the prevalence of fabricated feedback has become a significant concern, undermining the trustworthiness of these platforms. This project aims to develop a machine learning-based system for detecting fake reviews3 and assisting users in making informed choices. The Research study leverages a dataset consisting of labeled reviews, where each review is classified as fake or genuine1. This is used to train and judge the performance of a variety of machine learning models. The initial phase of the project involves preprocessing the textual data by applying techniques such as : • tokenization, • stop-word removal, and • stemming to extract relevant features. Furthermore, the system incorporates natural language processing (NLP) methods: • to capture semantic and syntactic information, 2 • enabling a depth knowing of the review content. 4 To increase the detection accuracy, the project explores the integration of additional contextual features, such as • User profiles, • Review timestamps, and • Review ratings. These above provided info for establishing patterns and identifying anomalies associated with fake reviews3.
U et al. (Fri,) studied this question.