Pengukuran Kepuasan Pengguna Mobile JKN pada Play Store Menggunakan Naïve Bayes Dan SVM

Abstract

Mobile JKN is a digital service developed by BPJS Kesehatan to facilitate public access to various healthcare administrative services online. As the number of users continues to increase, user reviews on Google Play Store have become an important source of information for evaluating service quality and user satisfaction. This study aims to analyze user sentiment toward Mobile JKN and compare the performance of Naïve Bayes and Support Vector Machine (SVM) algorithms for sentiment classification. The dataset was collected through a web scraping process from Google Play Store, resulting in 7,000 reviews. After data cleaning, 6,995 reviews were retained for analysis. The preprocessing stage included cleaning, case folding, tokenization, stopword removal, and stemming. Sentiment classification was then conducted into three categories: positive, neutral, and negative, using Naïve Bayes and SVM algorithms. The results show that the sentiment distribution consists of 60.76% positive, 17.08% neutral, and 22.16% negative reviews. SVM achieved the best performance with an accuracy of 92.10%, precision of 91.20%, recall of 90.50%, and F1-score of 90.80%, while Naïve Bayes achieved an accuracy of 89.50%, precision of 88.70%, recall of 87.90%, and F1-score of 88.30%. The findings indicate that users appreciate the convenience and accessibility of the services provided by Mobile JKN, although complaints related to login issues, OTP verification, and system stability remain prevalent. This study demonstrates that SVM outperforms Naïve Bayes in sentiment classification of Mobile JKN reviews and can provide valuable insights for evaluating and improving BPJS Kesehatan’s digital services.