Sentiment Analysis of the BCA Mobile Banking Application on User Reviews in the Google Play Store Using the Naïve Bayes Method

Abstract

The advancement of digital technology has encouraged banking services to shift toward more practical mobile banking, including the BCA Mobile application. This application is used by customers to perform financial transactions online. This study aims to analyze user sentiment toward the application using the Multinomial Naïve Bayes algorithm based on 1,000 Indonesian-language reviews taken from the Google Play Store. The analysis process includes text preprocessing, TF-IDF weighting, and model evaluation using accuracy, precision, recall, and F1-score metrics.The results show that the model performs well with an accuracy of 83.87%, precision of 92.31%, recall of 68.18%, and an F1-score of 78.36%. Most reviews indicate negative sentiment, mainly related to technical issues such as system errors and login difficulties, while positive reviews highlight ease of use and fast transactions. These findings are expected to provide input for developers to enhance the quality and stability of the BCA Mobile application.