Social Media Sentiment Analysis Using Machine Learning to Improve Digital Banking Services

Abstract

The development of digital banking services has transformed the interaction patterns between consumers and banking products or services, including at PT Bank Digital Nusantara, which operates the ABC application as its primary mobile-based financial transaction platform. This study aims to analyze public sentiment toward the Bank’s official social media content on Instagram and TikTok, as an effort to understand the perceptions, needs, and feedback, particularly from the bank’s customers. Using 10,198 comments collected during the period from August 31, 2024, to August 31, 2025, this study applies two Machine Learning methods, namely Naïve Bayes and Support Vector Machine (SVM), to classify customer sentiment and compare the performance of both models. The sentiment analysis results are then integrated into a SWOT analysis to identify strengths, weaknesses, opportunities, and threats, which are further examined through Root Cause Analysis using the Fishbone Diagram and 5 Whys approach to systematically determine the underlying causes of issues. The study shows that leveraging sentiment analysis plays a strategic role in supporting data-driven decision-making, enhancing the company’s responsiveness to service issues, strengthening brand awareness, and promoting the competitiveness of ABC by Nusantara as a digital bank in Indonesia. These findings are expected to contribute to the development of digital banking business strategies while enriching the academic literature on the application of sentiment analysis to improve service quality and customer experience.