SIGNAL App Review Sentiment Analysis using Support Vector Machine (SVM) on Google Play Store Comments

The SIGNAL (National Digital Samsat) application is a digital innovation that makes it easier to pay motor vehicle taxes in Indonesia. This study aims to analyze user sentiment towards the SIGNAL application through reviews on the Google Play Store, using Support Vector Machine (SVM) as a classification method. The analysis process includes the stages of review data collection, pre-processing (text cleaning, tokenization, stopword removal, and stemming), text transformation to numeric features using Term Frequency-Inverse Document Frequency (TF-IDF), and SVM model training. The dataset is taken from 10,000 of the latest reviews consisting of reviews classified into three sentiment categories: positive, negative, and neutral. The evaluation results show that the SVM model has a high accuracy of 91%, with consistent precision, recall, and F1-score values ​​in each sentiment category. Positive sentiment dominates reviews (59%), followed by negative sentiment (33.8%) and neutral (7.2%). This analysis provides valuable insights for developers to improve the quality of applications, especially in understanding user needs and expectations.