A Machine-Learning-Based Sentiment Classification Approach for Gojek Application
I Putu Edy Suardiyana Putra, Indra Hermawan, I Komang Arya Adi Kusuma, I Gede Deindra Dwija Puridiasta, Dewa Gede Bhaskara Pramudya, Clerencia Isabell Kowaas · 2024
This study proposes a machine-learning-based sentiment classification approach for Bahasa Indonesia, where Gojek App’s reviews are used for a case study. The sentiment classification is important since users often give 5-star ratings but leave negative reviews. A total of 19,772 reviews from The Google Play Store are used in this study. We use five machine learning algorithms to train the model, such as Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Naive Bayes (NB), and Stochastic Gradient Descent (SGD). The best results of our approach are 66.6% precision, 88.2% recall, and 69.2% F-score, where an SVM-based classifier is used. These results show a promising future for using machine-learning algorithms for sentiment classification.