<![CDATA[This study is motivated by the large number of customer reviews regarding perfume fragrances that are still evaluated manually, making it difficult to objectively identify the most preferred fragrance. This study aims to develop a sentiment analysis system to classify customer reviews into positive, neutral, and negative sentiments using the Naïve Bayes method and to determine the favorite perfume fragrance. The study employed a quantitative approach using sentiment analysis. The population consisted of all customer reviews at the perfume store under study, while the sample comprised reviews containing information about fragrance and review content. The research instrument consisted of customer review data collected through online questionnaires and direct interactions. Data analysis involved preprocessing, TF-IDF weighting, an 80:20 training-testing split, Naïve Bayes classification, and evaluation using a confusion matrix, accuracy, precision, recall, and F1-score. The results showed that the model achieved 71.43% accuracy on 42 testing data, with 30 correctly classified instances. Of the 36 reviews, positive sentiment accounted for 58.33%. CHRISTIAN DIOR - SAUVAGE obtained the highest positive sentiment, with 7 of 10 reviews or 70.00%. The conclusion indicates that Naïve Bayes can effectively be applied to classify customer sentiment and support the identification of favorite perfume fragrances for store decision-making]]>
Copyrights © 2026
