Analisis Sentimen Aplikasi Gojek Pada Ulasan Pengguna di Google Play Store Menggunakan Metode Support Vector Machine

Abstract

Gojek is an application that has garnered significant attention in Indonesia, offering a variety of services, including transportation, food delivery, and other on-demand services. This study aims to analyze user sentiment towards the Gojek app downloaded from the Play Store in 2024 using the Support Vector Machine (SVM) method. Data was collected through web scraping, comprising 8,000 reviews. The preprocessing steps included cleaning, case folding, tokenizing, stopword removal, stemming, and lexicon-based labeling, followed by TF-IDF and testing using a confusion matrix with Python. The sentiment labeling results revealed that the majority of reviews were negative at 44.25%, followed by positive sentiment at 36.62%, and neutral sentiment at 19.12%. The testing scheme applied 80% for training and 20% for testing. The analysis results showed an accuracy of 93.69%, recall of 93.66%, precision of 93.65%, and F1-Score of 93.67%. These results indicate that the SVM model is capable of classifying sentiment with a high level of accuracy. These findings can provide valuable insights for Gojek developers to enhance the app's service quality.