Sentiment Analysis Naive Bayes Method on SatuSehat Application

The SatuSehat application is an application that provides health services to users. This application is a development of the PeduliLindungi application which is used to handle vaccination history in the new normal era. Therefore, it is important to classify user reviews into positive and negative sentiments using the Naïve Bayes method. The use of this method because it can produce a model that is quite accurate and effective. The results of data collection in this study were 25,000 of which 18,359 were negative and 6,641 were positive. The results of the Naïve Bayes accuracy test are 97% with negative sentiment results, namely precision has a value of 98%, recall has a value of 98% and f1-score has a value of 98%, while positive sentiment results, namely precision has a value of 94%, recall has a value of 94 % and f1-score has a value of 94%. This study aims to classify user reviews of the SatuSehat application into positive and negative sentiments and assess the performance of the Naïve Bayes method regarding public opinion on the use of the SatuSehat application based on reviews from the Google Playstore application