Analisis Sentimen Pengguna Aplikasi Qur’an Kemenag Menggunakan Metode Support Vector Machine (SVM)

The Ministry of Religious Affairs' Quran application is designed to facilitate Muslims in reading and studying the Quran. This study aims to evaluate user sentiment towards the application based on reviews given on the Google Play Store. Although the application has a rating of 4.6 and more than 500 thousand downloads, there are still various complaints regarding bugs and performance. The method used in this study is Support Vector Machine (SVM) as the classification algorithm and Radial Basis Function (RBF) as the kernel. A total of 3,000 reviews were collected through web scraping techniques, then filtered by removing 3-star reviews considered ambiguous. After preprocessing, 2,677 reviews were obtained, which were then balanced using the Synthetic Minority Oversampling Technique (SMOTE). The data were labeled into two categories: positive and negative sentiment. The results showed that the classification accuracy before the implementation of SMOTE was 91% and increased to 97% after SMOTE was implemented. The results of this study provide a clear picture that although the Ministry of Religious Affairs' Quran application received a high rating, users still highlighted bugs, performance, and stability. With this sentiment mapping, developers can prioritize feature improvements and service quality enhancements to improve user experience and increase satisfaction.