Online consumer review platforms are becoming more important in shaping feedback for businesses and buying decisions for consumers, as seen by the growing use of these sites for electronic product reviews. A major obstacle, though, is the abundance of deceptive or fraudulent reviews that aim to artificially boost reputations and attract clients. This research focuses on the crucial task of detecting such fraudulent content by employing various machine learning classifiers on the Yelp dataset. The study’s main objective is to evaluate the effectiveness of these classifiers in identifying fake reviews. The primary goal of the research is to evaluate the effectiveness of these classifiers in identifying fake reviews. Results show that compared to other models, the Linear Support Vector Classification (SVC) model performs much better, with an accuracy of 87.9% using the Term Frequency-Reverse Document Frequency (TF-IDF) method and over 86.1% using the BoW feature. The proposed model shows that it can distinguish between genuine and fraudulent reviews by exceeding previous methods in terms of accuracy and F1-score. The proposed research demonstrates how important it is to use sophisticated machine learning methods to ensure the security of online review sites.