Sentiment classification on Amazon reviews using machine learning approaches

As online marketplaces have been popular during the past decades, the online sellers and merchants ask their purchasers to share their opinions about the products they have bought to help them grow their business.

Abstract

As online marketplaces have been popular during the past decades, the online sellers and merchants ask their purchasers to share their opinions about the products they have bought. As a result, millions of reviews are being generated daily which makes it difficult for a potential consumer to make a good decision on whether to buy the product. Analyzing this enormous amount of opinions is also hard and time consuming for product manufacturers. This thesis considers the problem of classifying reviews by their overall semantic (positive or negative). To conduct the study two different supervised machine learning techniques, SVM and Naïve Bayes, has been attempted on beauty products from Amazon. Their accuracies have then been compared. The results showed that the SVM approach outperforms the Naïve Bayes approach when the data set is bigger. However, both algorithms reached promising accuracies of at least 80%.