<![CDATA[Innovation in drug and food supervision is just one of many public service sectors boosted by the meteoric rise of Mobile applications, which is in turn driven by the public's demand for quick and efficient solutions and the pervasiveness of smartphones. In light of this need, the Republic of Indonesia's Food and Drug Supervisory Agency (BPOM) has released the BPOM Mobile app to facilitate public participation in the monitoring of food and drug products in circulation and to make information more easily accessible. Registered product information, breaking news, and the ability to submit complaints are all intended uses for this app. This research looked at the tone of BPOM Mobile reviews and discovered that most people were unhappy with the app, suggesting that it fell short of their expectations. This study utilized the Naive Bayes method in conjunction with the SMOTE Upsampling technique to assess sentiment. The accuracy, precision, and recall for the classification were 83.98%, 77.18%, and 96.49%, respectively. The results show that the Naive Bayes model with SMOTE does a good job of analyzing the sentiment of BPOM Mobile user reviews, and it also highlights the fact that the government needs to improve its application services. This study contributes in several aspects. First, this study presents a machine learning-based analysis to assess user satisfaction with public service applications. Second, the results of this study can be input for BPOM to improve the functionality and user experience in using the BPOM Mobile application.]]>
