Sentiment Analysis of Amazon Product Reviews Using Machine Learning and Deep Learning Models
Joy Chandra Gope, Tanjim Tabassum, Mir Md. Mabrur, Keping Yu, Mohammad Arifuzzaman · 2022
Due to the expansion of social networks and e-commerce websites, sentiment analysis or opinion mining has become a more active study issue in recent years. The objective of sentiment analysis is to identify and categorize the positive and negative sentiment expressed in a piece of text. Consumers can submit reviews with a specified rating on e-commerce websites like Amazon.com. As a result, in our paper, we sought to construct sentiment analysis related to product ratings and text reviews utilizing Amazon's dataset. Linear Support Vector Ma-chine, Random Forest, Multinomial Naive Bayes, Bernoulli Naive Bayes, and Logistic Regression were among the machine learning algorithms used. We acquired accuracy with the Random Forest classifier (91.90%). We also use RNN with LSTM as a deep learning approach in our paper and got maximum accuracy (97.52%). For our model RNN-LSTM is ideal approach.