At present the technological developments that affect each system and in the aspects of online transportation and transportation are currently the most popular transportation because of the ease of using these transportation services in mobile phone applications. Some people express their opinions and opinions about users of public transport through social media sites and other websites. This opinion can be used as a material for sentiment analysis to find out whether public transport services are positive or negative. The purpose of this study was to find out the sentiments in the tweet opinion and to find out the results of the classification of the Naive Bayes method, C4.5 and the Random Forest algorithm that were used and compared. In this study, from the results of testing with performance measurements the three algorithms use Cross Validation, Confusion Matrix and ROC Curve.At present the technological developments that affect each system and in the aspects of online transportation and transportation are currently the most popular transportation because of the ease of using these transportation services in mobile phone applications. Some people express their opinions and opinions about users of public transport through social media sites and other websites. This opinion can be used as a material for sentiment analysis to find out whether public transport services are positive or negative. The purpose of this study was to find out the sentiments in the tweet opinion and to find out the results of the classification of the Naive Bayes method, C4.5 and the Random Forest algorithm that were used and compared. In this study, from the results of testing with performance measurements the three algorithms use Cross Validation, Confusion Matrix and ROC Curve.