Analisis Sentimen Ulasan Dan Komentar Pengguna Pada Aplikasi Toco Menggunakan Algoritma Support Vector Machine

The development of digital applications has driven an increase in the number of user reviews and comments containing opinions about the services used, one of which is the Toco app. This study aims to automatically analyze the sentiment of Toco app user reviews and comments by grouping them into positive and negative categories. The method used is a Support Vector Machine (SVM) with Term Frequency–Inverse Document Frequency (TF-IDF) feature weighting. The
research data consisted of 3,000 reviews and comments collected from the Google Play Store, TikTok, and Instagram. They then underwent preprocessing stages including cleaning, case folding, tokenization, normalization, filtering, and stemming. The evaluation results showed that the SVM model achieved an accuracy of 93.17%, with a precision of 97.89%, a recall of 83.41%, and an F1-score of 90.07%. These result demonstrate that the TF-IDF-based SVM algorithm has good
and stable performance in classifying user sentiment towards the service quality of the Toco app.