Enhancing Mobile Banking Apps in Sri Lanka: UI or UX Classification on Negative Reviews

Abstract

The widespread adoption of mobile banking applications, initially accelerated by the COVID-19 pandemic, has permanently transformed financial transaction patterns. This study analyzed user reviews of popular mobile banking apps in Sri Lanka to evaluate user interface (UI) and user experience (UX) satisfaction. Using topic modelling techniques, we extracted and ranked significant UI and UX-related keywords from Google play store reviews respective to the banking application to identify prevalent user issues. The primary objective was to determine either UI or UX improvements were critical for future updates, improving user satisfaction and engagement. Classification algorithms including K-Nearest Neighbors (KNN), Random Forest, Extreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost), and Artificial Neural Networks (ANN) were used to classify UI and UX negative reviews based on topic-related keywords. The ANN model achieved 89% accuracy in classification, revealing that UI-related issues dominated negative feedback for a major government banking app. Future work will extend this analysis to other major Sri Lankan banking apps to provide comprehensive insights into specific UI and UX factors affecting user satisfaction.