Text Mining-Based Diagnosis and Improvement of UX Issues in Mobile Financial Applications

This study explores user experience issues in mobile financial applications by analyzing user reviews and identifying areas of satisfaction and dissatisfaction. Large-scale sentiment classification was performed using a model trained with practically labeled data, enabling efficient processing of over 130,000 user reviews. The analysis revealed key strengths such as intuitive navigation, smooth transactions, and clear information delivery, which suggest opportunities for improving personalization, UI consistency, and usability across devices. Conversely, common issues included authentication failures, unintuitive menus, and server instability, highlighting the need for simplified login processes, better interface structure, and technical reliability. Based on these insights, the study proposes actionable strategies to enhance user experience and outlines practical implications for data labeling and review analysis in UX diagnostics.