With the rapid development of the Internet, e-commerce platforms have become the main channel for people's daily shopping. Detailed product descriptions and user review systems play a key role in consumer purchase decisions. In this study, textual feature analysis is used as an assessment criterion, and whether the reviews are from customers is defined as a 0–1 dummy variable, which is used as the dependent variable. We use a decision tree classification model to train the data and a cross-validation method for supervised learning. In order to visually compare the classification effect of the models, we choose the support vector machine (SVM) classification model and the random forest (RF) classification model. The results show that the classification results of decision tree classification and RF classification on the validation set are consistent, which further validates the reasonableness of decision tree classification. Ultimately, practical suggestions for consumers to shop online were provided based on the model results. This study contributes to an in-depth understanding of consumers’ shopping decision-making process on e-commerce platforms and provides useful insights for improving the shopping experience and online shopping decisions.