Jakarta Kini (JAKI) is a super-app developed by Jakarta Smart City to simplify public service access in the DKI Jakarta Province. As an application widely used by the public, JAKI has received thousands of user reviews on the Google Play Store, reflecting public opinion on its features and performance. This study aims to classify user reviews into positive, negative, and neutral sentiment categories using the Naive Bayes algorithm. The research method includes collecting review data through web scraping, text preprocessing, sentiment analysis, data labeling, model building, and model evaluation using the Orange platform. The results show that the Naive Bayes algorithm successfully classified 3,226 review data with a perfect accuracy of 100%, as confirmed by the confusion matrix and other evaluation metrics (precision, recall, F1-score, MCC). The sentiment distribution reveals that most reviews are neutral, followed by negative and then positive sentiments. This indicates that public perception of the JAKI application tends to be moderate, highlighting the need for developers to improve the quality of digital public services. This research is expected to serve as a reference for utilizing machine learning-based sentiment analysis in evaluating public service applications
Ridwan Wicaksono (Wed,) studied this question.