Sentiment Analysis of Amazon Products Using Ensemble Machine Learning Algorithm

Abstract

In recent years, Sentimental Analysis is used in all online product firms. The number of users using the particular product has increased which makes the industry to improvise the characteristics of the product. These days, many users who are using websites, blogs, online shopping tends to review the products they used. These reviews were taken into consideration by other users during their search for products. Hence the industry has found the root of delivering the correct product searched by the user based on the reviews of the users using the concept of sentimental analysis. Sentimental Analysis is the concept of data analysis where the collections of reviews are taken into consideration, and those reviews are analyzed, processed and recommended to the user. The reviews given are longer and which consist of a few paragraphs of content. In this paper, the dataset is collected from the official product sites. Initially, these reviews must be pre-processed in order to remove the unwanted data’s such as stop words, be verbs, punctuations, and conjunctions. Once, the pre-processing is over the trained dataset is classified using Naive Bayes and SVM algorithm. These existing algorithms provided the accuracy which is not worth enough. Hence, an ensemble approach has been applied to enhance the accuracy of the given reviews. An ensemble is a classification approach by combining two or more algorithms and calculate the mode value based on the vote reference for every algorithm which is used. In this paper, Naive Bayes, SVM, and Ensemble algorithm are combined. We proposed an Ensemble method that helps in providing better accuracy than the current existing algorithm. Once the accuracy is calculated, based on the reviews, the particular product is recommended for the user.