Amazon product reviews sentiment analysis using supervised learning algorithms

Abstract

E-commerce is gaining traction in today’s digitalized environment by taking products closer to customers without forcing them to leave their homes. A customer must study hundreds of reviews before making a purchase. The amount of internet evaluations for a single product can easily approach millions and make tracking and understanding of client feedback difficult. In the era of machine learning, however, it would be much easier to gain thousands of input and knowledge from them if a model were employed to polarize and understand from them. Consequently, sentiment analysis is a new study area combining natural language processing and text analytic to extract subjective information from sources and classify the polarity of expressed sentiments. We have employed Vector Machine Support, Naive Bays, Decision Tree, Random Forest, Logistic Regression, and MLP Classifiers for large-scale supervised education on the Amazon dataset and obtained satisfactory results. In the meantime, the MLP classifier produced the best results. Finally, this paper discusses sentiment analysis and product feedback opinion mining.