<![CDATA[The development of digital technology has encouraged banks to provide application-based financial services, one of which is BYOND by BSI, which carries the concept of Islamic banking. However, various technical obstacles such as service disruptions and application errors in using this application have caused dissatisfaction among users. Therefore, sentiment analysis is needed to understand user responses comprehensively. This study aims to classify user sentiment towards the BYOND by BSI application by utilizing the Support Vector Machine (SVM) and Random Forest algorithms. The data used are 35,000 user reviews collected from the Google Play Store through crawling techniques, then automatically labeled using a rule-based method based on rating values. The analysis process was carried out using the SEMMA approach, which includes the stages of text cleaning, word weighting using TF-IDF, and dividing the data into 80% training data and 20% test data. The test results showed that the SVM algorithm had the best performance with an accuracy of 93.16%, while Random Forest obtained an accuracy of 90.33%. The majority of the analyzed reviews showed negative sentiment. These findings are expected to provide input in improving the quality of the BYOND by BSI application service.]]>
Copyrights © 2025
