<![CDATA[The rapid growth of mobile commerce has positioned mobile applications as critical touchpoints influencing customer satisfaction and business performance. This study aims to identify customer satisfaction patterns by analyzing negative user reviews of the JIWA+ mobile application using the Support Vector Machine (SVM) algorithm. A total of 1,025 reviews collected from the Google Play Store during the period 2022–2025 were processed through text preprocessing, TF-IDF feature extraction, and sentiment classification into positive, neutral, and negative categories. The SVM model achieved an overall accuracy of 0.746, demonstrating reliable capability in classifying sentiment polarity, particularly in detecting negative reviews. The findings indicate that 35.7% of reviews reflect negative sentiment, highlighting significant dissatisfaction among users. The dominant complaint themes include transaction failures (“pesan”), feature usability issues (“pakai”), and discrepancies between digital information and outlet conditions (“gerai”). These issues primarily relate to system reliability, payment functionality, and digital–offline integration. From a business management perspective, this study positions sentiment analysis as a strategic analytical tool that transforms unstructured customer feedback into actionable managerial insights. The results contribute to the literature on mobile applications, sentiment analysis, and customer satisfaction by demonstrating how machine learning–based approaches can support data-driven decision-making in enhancing digital service performance and sustaining competitive advantage.]]>
Copyrights © 2026
