Sentiment Analysis Based on Online Women's Clothing Reviews and Ratings Using Machine Learning Approaches

Abstract

Analyzing how customers to act is essential for e-commerce marketing strategies because e-commerce can significantly boost economic growth. In natural language processing (NLP), techniques like sentiment analysis are used to determine whether data is positive, negative, or neutral. Using the user reviews in our database, we can build a machine-learning model that gives the right sentiment for each product. In addition to helping customers understand the product better, an accurate sentiment research also helps the business gain better market feedback. In this study, we perform sentiment analysis on a data set from online reviews of women's clothes downloaded from Kaggle. Three well-known machine learning algorithms were used to tackle the issue: logistic regression, Naive Bayes classifiers, and Support Vector Machine (SVM). The best results came from the LR algorithm, which had the best AUC value and accuracy.