The rapid development of digital transportation, such as Gojek, requires a deep understanding of user satisfaction. This study analyzes the sentiment of Gojek application reviews to evaluate public opinion and compare the performance of the Support Vector Machine (SVM) and Random Forest models. A quantitative experimental method was applied to 30,055 user reviews for versions "4" and "5" from the Google Play Store. The data underwent comprehensive text preprocessing, automatic sentiment labeling using VADER enriched with an Indonesian lexicon, and TF-IDF feature extraction. The training data imbalance was addressed using SMOTE before the data was split for training and testing. The results show that user sentiment was dominated by positive (38.9%) and neutral (38.2%) categories. In the performance evaluation, the SVM model demonstrated superior performance with 96% accuracy and an F1-score of 0.96, outperforming the Random Forest model, which achieved 93% accuracy and an F1-score of 0.93. In conclusion, SVM is a more effective model for sentiment classification of Gojek reviews. Future research is recommended to refine the lexicon and implement aspect-based analysis to obtain more detailed insights.