Sentiment Analysis of Product Reviews for E-Commerce Site Using Machine Learning

Sentiment Analysis of Product Reviews for E-Commerce Site Using Machine Learning

Om  Sudokar¹, Vitthal Khandagale², Komal Ware³, Farhan Quazi⁴, Prof. K. R. Hole⁵

¹Scholar of Computer Science and Engineering, Prof. Ram Meghe Institute of Technology and Research Badnera-Amravati

²Scholar of Computer Science and Engineering, Prof. Ram Meghe Institute of Technology and Research Badnera-Amravati

³Scholar of Computer Science and Engineering, Prof. Ram Meghe Institute of Technology and Research Badnera-Amravati

⁴Scholar of Computer Science and Engineering, Prof. Ram Meghe Institute of Technology and Research Badnera-Amravati

⁵Assistant Professor of  Computer Science and Engineering, Prof. Ram Meghe Institute of Technology and Research Badnera-Amravati

Abstract: A large majority of users of e-commerce platforms communicate in English. The corporate environment of today is highly competitive,, sentiment analysis is widely used in the e-commerce industry to improve efficiency and better understand to make business decisions. we have developed a machine learning model where reviews on three English language are used and applied two machine learning algorithms. As part of this study, we examined specific accuracy, precision, recall, F1 scores, and ROC area and compared them to prior studies. We have prepared dataset and labeled all the reviews data as Negative, Positive, Neutral sentiment. Using machine learning techniques, the preprocessed dataset was trained, and the model's performance was evaluated.. For the English dataset, Random Forest algorithm performed best by achieving  94% accuracy.

Keywords: Sentiment Analysis, E-commerce Review, Review, Machine Learning