<![CDATA[This research analyzes the sentiment of JKN mobile app reviews on Google PlayStore using the IndoBERT model, a deep learning-based language model designed for Indonesian text. The research process involved review data collection, text pre-processing, and sentiment classification into three categories: positive, negative, and neutral. The results show that the model performs very well, with an average accuracy of 97.28% and best metrics of 98.27% on accuracy, precision, recall, and F1 score. The specific contribution of this research is the development of a deep learning-based approach for sentiment analysis of Indonesian texts, particularly in the health sector through mobile applications. The findings not only provide insight into user perceptions of the JKN app, but also provide a basis for feature improvements and service enhancements. The implications of this research can support developers in designing strategies to improve the quality of digital-based health services in Indonesia.]]>
Copyrights © 2025
