An Intelligent Sentiment Analysis System For Product Reviews Using Machine Learning

DOI

https://www.doi.org/10.59256/ijsreat.20260602027

Published In

Volume 6, Issue 2 (2026)

Published Online

March-April 2026

Pages

220-226

Article Type

Original Article

Check for updates

With the rapid growth of e-commerce platforms, analyzing customer opinions has become essential for businesses to understanduser satisfaction and improve product quality. This paper presents “Sentira Enterprise”, an intelligent sentiment analysis system designed to classify product reviews into positive, negative, and neutral categories.The proposed system utilizes Natural Language Processing (NLP) techniques, including text preprocessing, tokenization, and feature extraction using Term Frequency-Inverse Document Frequency (TF-IDF). A machine learning model based on the Naive Bayes algorithm is implemented to perform sentiment classification with optimized accuracy and performance.In addition to text-based analysis, the system supports bulk review processing through CSV uploads and integrates audio-based sentiment analysis using speech-to-text conversion. The system is deployed using a Streamlit-based user interface, providing real-time predictions, visual dashboards, and downloadable PDF reports.Experimental results demonstrate that the model achieves efficient and accurate sentiment classification while maintaining scalability for large datasets. The proposed system offers a practical and user-friendly solution for businesses to gain actionable insights from customer feedback