Analisis Sentimen Ulasan Aplikasi Signal Pada Google Play Store Menggunakan Metode Support Vector Machine

Currently, an application that can be used on a Smartphone called SIGNAL has emerged. The SIGNAL application is the SAMSAT DIGITAL NASIONAL application, an application that is useful for making it easier for the public to pay motor vehicle taxes safely and easily. SIGNAL application users have increased over time, because more and more people know about the existence of the application. The public uses the SIGNAL - SAMSAT DIGITAL NASIONAL application by downloading it via the Google Play Store and telling their experiences in the comments on the Google Play Store. The purpose of this study was to implement sentiment analysis of reviews from the Google Play Store on user comments on the SIGNAL - SAMSAT DIGITAL NASIONAL application using the Support Vector Machine method. 2. Data was obtained from the results of scraping on the Google Play Store with a total of 1000 data. The testing method was carried out using a confusion matrix. From the tests that have been carried out, the results of the research from the sentiment analysis of reviews on the SIGNAL - SAMSAT DIGITAL NASIONAL application using the Support Vector Machine algorithm obtained the best results in the data division method with a ratio of 90:10 with the SMOTE technique producing an accuracy of 0.95. Meanwhile, the worst test was on the data division method with a ratio of 70:30 which produced an accuracy of 0.85.