AN EMPIRICAL STUDY ON FAKE REVIEW DETECTION

In recent years, the Internet has opened up opportunities for manufacturers and retailers to advertise and sell their products online. Online shopping is becoming a habit of consumers. Although there are many benefits of buying and selling online, such as easy product selection and comparison from many different sellers before deciding which one to buy, reading comments before buying a product is a habit of customers. It helps them learn from the experiences of former buyers. However, buying based on product reviews is risky, especially fake reviews. These reviews affect the buyers’ purchase decisions. Detecting fake reviews is a critical problem. This study proposed a machine learning-based framework for detecting fake reviews by extracting features from text and deployed six machine-learning models for classification tasks. Experimental results showed that the SVC is a reliable machine-learning algorithm for classifying truthful reviews and fake reviews using the TF-IDF feature extraction technique.