Life and internet technology have always been inextricably linked. Not only does it make people’s lives easier, but it also enables information sharing, which is especially important in e-commerce. People leave messages and express their emotions online. Accurate classification enables customers to understand products better and provides companies with more precise market insights. This paper presents a new method for analyzing and classifying women’s online ratings based on review data. The classification utilizes a publicly available dataset from Kaggle, which contains reviews of women’s clothing obtained from an online retail platform. The CRISP-DM methodology is employed to provide a systematic framework for the work. Four machine learning algorithms were implemented and evaluated: Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), and Naive Bayes (NB). The dataset was divided into five distinct training-testing splits to assess the algorithms across different data distributions. Experimental results have shown that logistic regression (LR) outperforms other algorithms, achieving an accuracy metric of 96% based on a 50-50 split for training and testing data, respectively.