UNDERSTANDING USER SATISFACTION THROUGH SENTIMENT AND FEATURE-BASED SENTIMENT ANALYSIS: A COMPARATIVE STUDY OF GRAB AND UBER RIDE-HAILING SUPER APP REVIEWS ON GOOGLE PLAY STORE USING PYTHON AND RAPIDMINER

User satisfaction is defined as the comfortability or contentment of the end-users of an application (Trymata & Trymata, 2024). This study investigates user satisfaction with two major ride-hailing applications, Grab and Uber, by analyzing 1,000 English-language Google Play Store reviews for each app, written in December 2024. Using Python and RapidMiner, sentiment analysis was conducted with the VADER technique to classify reviews as positive, neutral, or negative. Word clouds of positive and negative reviews for both applications were also done to reveal the common themes in user reviews. Feature-based sentiment analysis was also applied to evaluate user feedback on specific app features, with results visualized through bar charts and tables. Findings revealed that Uber received more positive reviews and fewer complaints than Grab. While users valued Grab’s multifunctionality and affordability, they frequently reported technical issues and poor customer service. Uber was praised for its performance, interface, and professional drivers, though concerns about pricing and availability were noted. The study concludes that user satisfaction is shaped by reliability, responsiveness, and app performance.