<![CDATA[Portal Pulsa application is a pulse refilling and bill payment service platform that has a high number of reviews on the Google Play Store. These reviews can be used to determine user perception of application services. This study aims to perform sentiment analysis on Portal Pulsa application user reviews using machine learning methods. The research stages include collecting review data from the Google Play Store, text preprocessing, feature extraction using TF-IDF, and sentiment classification using several machine learning algorithms, namely Naïve Bayes, Linear Support Vector Machine (Linear SVM), Random Forest, K-Nearest Neighbor (KNN), and Decision Tree. Model evaluation was performed using accuracy, precision, recall, and F1-score metrics. The results showed that the Linear SVM algorithm provided the best performance with an accuracy value of 93.24%. These results indicate that Linear SVM is effectively used in classifying the sentiment of Portal Pulsa application user reviews.]]>
Copyrights © 2026
