<![CDATA[Understanding customer satisfaction is essential for digital service providers operating in highly competitive markets, particularly in the online transportation sector, as numerical ratings alone are often insufficient to capture the complexity of user experiences expressed in textual reviews. This study employs Aspect-Based Sentiment Analysis (ABSA) to obtain a more granular understanding of customer satisfaction based on user reviews of the Aku Cinta Indonesia (ACI) application collected from the Google Play Store. Five key service aspects are examined, namely User Experience, Payment, Service, Promo and Benefit, and Security and Access. To support the analysis, two sentiment classification approaches, Naive Bayes and IndoBERT, are utilized to evaluate sentiment polarity at the aspect level. The results indicate that User Experience, Payment, and Service are the most influential aspects shaping customer satisfaction, as they consistently appear in both positive and negative reviews. While both models demonstrate reliable performance, IndoBERT achieves higher classification accuracy at 82% compared to Naive Bayes at 77%, indicating a stronger ability to capture contextual nuances in Indonesian user generated content. From a business perspective, these findings highlight how ABSA transforms unstructured customer feedback into actionable insights that enable service providers to identify critical improvement areas, prioritize service quality enhancements, and strengthen customer satisfaction strategies. This study demonstrates the value of ABSA as a business analytics tool that supports data driven decision making and enhances competitiveness in the digital transportation service market.]]>
Copyrights © 2026
