Treffer: Sentiment Analysis of Triv Application Reviews using Support Vector Machine Algorithm

The growing popularity of the Triv application as a cryptocurrency transaction platform in Indonesia has generated various user reviews that reflect perceptions of service quality. This study focuses on exploring user opinions through sentiment analysis techniques employing a classification approach based on the Support Vector Machine (SVM) algorithm. The data, sourced from user reviews on the Google Play Store, is analyzed through a series of systematic stages, including sentiment labeling, text preprocessing, feature extraction, model construction, and performance evaluation of the resulting classifier. The experimental results show that SVM can accurately identify sentiment polarity, achieving an accuracy rate of 96%. These findings highlight the potential of machine learning approaches in understanding user perceptions of digital financial applications.