Sentiment Analysis of Fintech Application User Reviews using the CRISP-DM Framework for Product Development Prioritization

The rapid growth of fintech applications has increased the need for sentiment analysis to understand user perceptions of the offered products. This study focuses on sentiment analysis of user reviews for the Flip application on Google Play Store by applying the Support Vector Machine (SVM) algorithm within the CRISP-DM framework. The analysis process involves text preprocessing, sentiment labeling using a pretrained BERT model, and classification using SVM with TF-IDF feature extraction. The results indicate that the majority of users express positive sentiment (56.9%), primarily regarding cost efficiency, transaction ease, and product speed. However, negative sentiment (43.1%) is also present, mainly concerning additional fees, transaction delays, and technical issues in app usage. A topic modeling analysis using the Latent Dirichlet Allocation (LDA) method identifies key topics that highlight both Flip's strengths and challenges. The findings suggest that while Flip holds significant potential in meeting user needs, improvements are needed in product aspects, cost transparency, and app performance optimization. This study is expected to serve as a strategic foundation for fintech app developers to enhance data-driven product quality, ultimately increasing user satisfaction and loyalty.