<![CDATA[Student satisfaction reflects educational quality, influences retention, and enhances institutional reputation. This study examines the impact of student performance and motivation in online learning using Naïve Bayes and Logistic Regression. Data from 316 respondents at PLN Institute of Technology, collected during the COVID-19 pandemic via Microsoft Teams, were divided into 80% training and 20% testing. The process included questionnaire distribution, data labeling, parameter determination, and normalization to ensure completeness and reliability. Questionnaire data is stored in Excel format and was processed using Python for programming, Pandas for data manipulation, and Kaggle for dataset management, before being analyzed with Naïve Bayes and Logistic Regression. Finally, the processed data is tested for accuracy using confusion matrix. The results show high precision, recall, f1-score, and accuracy for both methods, with Naïve Bayes achieving an accuracy 93.75% to 97.44% and Logistic Regression achieving 98.95%. In summary, Naïve Bayes can be optimized with threshold adjustments, but Logistic Regression is more reliable than consistent, maintaining high accuracy across different thresholds. Institutions can update their strategies using the latest data to enhance learning experiences. From those results, it can be concluded that Naïve Bayes method should be enhanced, while Logistic Regression is proven reliable. In the future, researchers are encouraged to use more diverse datasets while also considering external factors such as technological infrastructure and psychological support.]]>
