A Study On Identifying Fake Reviews by Comparative Analysis of Supervised Machine Learning Algorithms

Abstract

In today’s time e-commerce sites are playing a vital role. Customers are attracted to those product companies whose reviews and quality are good. A positive review can bring fame to the company but on the other hand, a negative review can bring defame to the company. The review section which we are talking about can be fake sometimes. This fake review is harmful to both seller and buyer. A lot of classifiers are used to detect these fake reviews. So our work will be to check the best classifier through their accuracy for detection. Our goal is to detect fake reviews which is a text classification issue. We gathered our data, preprocessed the dataset, feature extraction, and apply supervised machine learning algorithms in the dataset. So, our main target area is to develop a model which classifies a given text either fake or real review, and also do the comparison between different classifiers.