<![CDATA[Significant transformations in the user interface and features of the BRImo application throughout 2024 have generated diverse responses from users. Monitoring changes in customer perceptions is essential for developers to evaluate the effectiveness of these updates and maintain the quality of digital banking services. This study aims to compare user sentiment before and after the BRImo application update using a text mining approach with the Support Vector Machine (SVM) algorithm. User reviews were collected from the Google Play Store through a web scraping technique. The collected data were processed through several text preprocessing stages, including cleaning, case folding, tokenization, stopword removal, and stemming. Furthermore, the Term Frequency–Inverse Document Frequency (TF-IDF) method was applied for feature weighting before classification using the SVM algorithm. The experimental results show that the SVM model achieved an accuracy, precision, recall, and F1-score of 92%, indicating its effectiveness in sentiment classification. Comparative analysis revealed an increase in negative sentiment after the application update, mainly related to login issues, authentication problems, adaptation to the new interface, and system stability. In contrast, positive sentiment remained associated with the application's comprehensive features and transaction convenience. In conclusion, technical stability after system updates has a significant influence on user satisfaction, while the SVM algorithm provides an effective automated approach for evaluating user feedback and supporting future application improvements.]]>
