<![CDATA[User reviews on the Google Play Store serve as a primary source of information for assessing the quality and satisfaction levels of an application. As expectations for game quality and realism continue to rise, Bus Simulator Indonesia still faces a range of technical issues — including bugs, glitches, visual disturbances, and suboptimal controls — that directly impact user comfort and overall experience. This study aims to classify the sentiments expressed by Bus Simulator Indonesia users into three categories: positive, neutral, and negative. In addition, it compares the performance of three machine learning algorithms: Naive Bayes, Support Vector Machine (SVM), and Logistic Regression. The research pipeline consists of data collection, text preprocessing, word weighting, modeling, and model evaluation. The dataset comprises 19,482 Indonesian-language reviews gathered from the Google Play Store between 2021 and 2025, evaluated across three data-splitting scenarios: 90:10, 80:20, and 70:30. The results show that Support Vector Machine consistently achieved the highest accuracy across all scenarios, recording 84.3%, 82.0%, and 81.0%, with weighted F1-score values of 0.83, 0.82, and 0.81, respectively, indicating a more stable classification performance. Logistic Regression followed with accuracies of 81.1%, 79.9%, and 79.0%, with weighted F1-score values of 0.79, 0.78, and 0.77, respectively, while Naive Bayes produced the lowest scores overall. These findings confirm that Support Vector Machine is the most effective algorithm for sentiment classification of Bus Simulator Indonesia user reviews.]]>
Copyrights © 2026
