Sentiment Analysis of Public Service Performance in Jakarta Using the Naïve Bayes Algorithm and Support Vector Machine

An agency's service performance is crucial. Community feedback is one source of information that the government needs in order to enhance its performance. Nowadays, people frequently utilize social media to voice their dissatisfaction with the services they have received, offer recommendations for work programs, or just to get the most recent information. The outcomes of the public's opinions can be utilized as assessment material by business owners or associated agencies in order to make changes and raise the standard of future performance. The public can use a variety of social media platforms in Jakarta to voice any grievances, requests for information, and suggestions about the city's development process. Instagram is one of the social networking platforms that users frequently utilize. Monitoring every Instagram account on every Instagram account is a difficult task to complete by hand. Public opinion found on social media is automatically categorized in this study. Support Vector Machine (SVM) and the Naïve Bayes method are used for classification. Instagram user @dkijakarta provided the data that was used. After processing, the data will be divided into three sentiment classes: neutral, negative, and positive