<![CDATA[Skincare products play an important role in personal care and consumer satisfaction. However, the wide variety of products and price ranges often makes it difficult for consumers to assess whether a product offers good Value for Money. This study proposes a machine learning approach to evaluate the Value for Money of skincare products using customer reviews from the Sephora platform. Review texts were preprocessed using Natural Language Processing (NLP) techniques, including lowercasing, tokenization, stopword removal, and lemmatization. Text features were extracted with TF-IDF and combined with product price, helpfulness score, and sentiment score. Support Vector Machine (SVM) and Logistic Regression were compared for classification, while SMOTE was applied to address class imbalance. Both models achieved similar performance, with an accuracy of 0.76, a macro-average F1-score of 0.70, and a weighted-average F1-score of 0.78. The results indicate that combining customer reviews with product-related features can effectively assess Value for Money and support consumers in making better purchasing decisions.]]>
Copyrights © 2026
