SENTIMENT ANALYSIS OF JIWA+ APPLICATION USERS BASED ON MACHINE LEARNING FOR DIGITAL SERVICE EVALUATION

Abstract

The rapid growth of Indonesia’s coffee industry has increased competition and encouraged coffee businesses to adopt digital technologies to improve service quality. Janji Jiwa utilizes the Jiwa+ mobile application to support digital services; however, user reviews on the Google Play Store show mixed sentiments toward the application. This study aims to analyze user sentiment toward the Jiwa+ application using machine learning based sentiment analysis as a tool for evaluating digital service quality in the coffee business sector. User review data were collected from the Google Play Store through web scraping and processed using text preprocessing techniques, including cleaning, case folding, tokenization, normalization, stopword removal, and stemming. The reviews were manually labeled into positive and negative sentiments and classified using Support Vector Machine (SVM) and Naive Bayes algorithms with TF-IDF feature extraction. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. The results indicate that SVM outperforms Naive Bayes, achieving an accuracy of 92% compared to 89% for Naive Bayes. Feature analysis shows that positive sentiment is associated with ease of use and promotional benefits, while negative sentiment highlights issues related to payment processes, application reliability, and stock availability. These findings demonstrate that sentiment analysis can provide valuable insights for improving digital services in the coffee industry.