<![CDATA[JKN Mobile application has become one of the main means for Indonesians to access health services provided by the National Health Insurance (JKN) program. In this context, sentiment analysis of user reviews of the JKN Mobile application is important to understand user perceptions and experiences. This research uses the SVM classification method to carry out in-depth sentiment analysis of user reviews of the JKN Mobile application. Research steps include collecting user review data, data preprocessing, feature extraction, data sharing, SVM model training, and model performance evaluation. User review data is collected and processed, then trained using an SVM model to classify review sentiment into positive, negative, or neutral. This research tested the level of accuracy by comparing the distribution of training data with testing data. The aim of this method is to explore various optimization techniques to increase the accuracy of the SVM model in classifying the sentiment of user reviews of the JKN Mobile application. This can include using optimization algorithms such as Particle Swarm Optimization (PSO) to find the best parameters for the SVM model. The research results show that the SVM accuracy level after implementing PSO or Grid Search experienced an increase in accuracy, initially the SVM accuracy level was 81%, after optimization, the accuracy level obtained a higher accuracy of 85%.]]>
Copyrights © 2024
