<![CDATA[The rapid growth of mobile-based ride-hailing services in Indonesia has significantly increased the use of digital platforms such as Gojek. This growth has generated a large volume of user reviews on the Google Play Store, reflecting user experiences, perceptions, and satisfaction levels. However, manual analysis of these reviews is inefficient due to their volume and subjective nature. Therefore, this study aims to model sentiment patterns in Indonesian Gojek user reviews using a machine learning approach. This research applies the Support Vector Machine (SVM) algorithm combined with Term Frequency–Inverse Document Frequency (TF-IDF) for feature extraction. The dataset consists of 5,000 Indonesian-language user reviews collected from the Google Play Store. Text preprocessing includes case folding, cleaning, tokenizing, stopword removal, and stemming. Sentiment labeling is performed automatically using a lexicon-based approach, classifying the data into positive, negative, and neutral categories. The results indicate that neutral sentiment dominates the dataset (60.26%), followed by positive (31.34%) and negative (8.40%) sentiments. The SVM model achieves an accuracy of 72.58%. The neutral class shows the highest recall (99.17%), while the positive class achieves the highest precision (99.65%). These findings indicate that TF-IDF and SVM are effective for modeling sentiment patterns, although challenges remain in handling ambiguous expressions, such as sarcasm or mixed sentiments, which can lead to misclassification.]]>
Copyrights © 2026
