Mobile banking is very competitive after becoming a form of innovation in the banking sector in the context of financial technology. The purpose of this study is to analyze the performance of mobile banking using bank user reviews through text mining. This study used the Knowledge Discovery in Database (KDD) method to compare the performance of 4 mobile banking in Indonesia for the January-September 2022 period. The modeling carried out is Machine Learning (ML) through the Naïve Bayes Classifier Algorithm. This study found three findings related to the performance of mobile banking. First, the highest accuracy value obtained from the modeling results of the Naïve Bayes Algorithm was 90% on BRImo. Second, the highest positive sentiment was 54.8% on Livin' by Mandiri and the highest negative sentiment was 49% on BCA Mobile. Third, generally, the positive sentiment produced more 47% than the negative sentiment 40%. This study suggests that banking companies have to make improvements to the performance and system of the mobile banking application based on the results of review evaluations in accordance with customer needs.