An Analysis of Sentiment in E-Commerce Products Review: A Detailed Survey on Methods

Abstract

Review sites for different products and services have increased due to the development of several e-commerce websites. Nowadays, reviews are a simple way for people to learn more about the goods and services they intend to utilize. Here, categorizing the polarity of product reviews is a crucial function of sentiment analysis (SA). Today, sentiment research allows businesses to comprehend consumer attitudes towards goods and services in a worldwide economy. Negative statements significantly impact sentiment identification.A sentiment analysis that merely provides overall polarity when there are many reviews must be more comprehensive. Finding reviews of the product’s particular parts (features) will be challenging. This study aimed to examine machine learning (ML) and deep learning (DL) models for predicting customer sentiments in product evaluations on e-commerce sites. The perspectives on the procedure, necessary activities, and tactics of SA that have been highlighted in many studies are presented in this study. Additionally, different difficulties with the sentiment classification process are discussed.