<![CDATA[As the largest public transportation system in Jakarta, TransJakarta frequently becomes a topic of discussion on Twitter due to various user experiences ranging from satisfaction to service-related complaints. The abundance of public opinions provides an opportunity to apply machine learning–based sentiment analysis as a faster, more objective, and measurable method for evaluating service quality. This study employs the Naive Bayes algorithm to classify tweet sentiments into positive, negative, and neutral categories. Data were collected through a crawling process using keywords related to TransJakarta, then processed through several stages including tokenization, text cleaning, stopword removal, and stemming to produce analysis-ready data. Several service aspects discussed most frequently include bus arrival punctuality, travel comfort, and fleet conditions. The analysis results indicate that the Naive Bayes model can accurately identify public sentiment patterns in near real-time, enabling the detection of opinion trends across different service dimensions. These findings demonstrate that Naive Bayes–based sentiment analysis can serve as an effective tool for monitoring public perception and provide a strong foundation for TransJakarta in formulating strategies to improve service quality.]]>
Copyrights © 2026
