Analysis of Customer Reviews Using Opinion Mining

Abstract

Since 2012, internet usage in the world has grown from a mere 1.8% to 59% in 2020. The numbers of online shopping webs in the world are becoming greater than ever before. People use the internet to access or update reviews. Many reviews are long and a few sentences contain an opinion on the products. Each product can get hundreds or thousands of customer reviews. Opinion mining is widely used in reviews and social media for a variety of applications, ranging from marketing to customer service. This paper is a system of opinion mining or sentiment analysis on the level of user satisfaction of product reviews. Opinion mining works for finding and classifying opinions in customer reviews on any products as either positive or negative. Support Vector Machine algorithm with RBF (Radial Basis Function) kernel is used to classify reviews; this is one of the supervised opinion mining techniques. The classification result obtained 87% accuracy using the camera product reviews dataset from amazon.com.