


Volume 20 No 20 (2022)
Download PDF
An Efficient Machine Learning-Based Sentiment Analysis for Amazon Product Reviews
K.Sravana Kumari , Dr. B. Manjula
Abstract
Web portals such as Amazon collect a large quantity of consumer input regularly. It might be a time-consuming and
tedious task to read through all of the criticism. They must classify people's thoughts as they are presented in feedback forums. A
feedback management system is an example of one application that could benefit from this. They classify each comment and
review according to one of these classes and use these classes to inform our purpose of an overall rating for the product. It is
necessary for the company to have a comprehensive picture of the feedback provided by customers and to focus its attention on
the appropriate areas. This results in an increase in the number of customers who are loyal to the company, as well as a growth in
business, reputation, and the value of the brand, as well as profits. As a result, they suggest that we use the aspect terms, also
referred to as the targets from the texts, to extract the users' feelings regarding particular qualities associated with the products.
This study built a model to predict the comment's sentiment based on the comment declaration using Python and Support vector
machine (SVM), random forest (RF), and Improved logistic regression (ILR) are the three different machine learning techniques.
The dataset used in this study was compiled from customer reviews of musical instruments sold on Amazon.com. SMOTE is
utilized so that the unbalanced dataset can be managed, and AUC and ROC are used so that the optimal approach can be
determined. For the review text's classification, the solution they have given is based on a logistic regression that uses the Kernel
Density Estimation approach. Classification evaluation indicators like F1-score, precision, recall, and accuracy are utilized in the
evaluation process.
Keywords
Sentiment analysis, Text classification, kernel function, logistic regression
Copyright
Copyright © Neuroquantology
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Articles published in the Neuroquantology are available under Creative Commons Attribution Non-Commercial No Derivatives Licence (CC BY-NC-ND 4.0). Authors retain copyright in their work and grant IJECSE right of first publication under CC BY-NC-ND 4.0. Users have the right to read, download, copy, distribute, print, search, or link to the full texts of articles in this journal, and to use them for any other lawful purpose.