<![CDATA[Burnout among university students is a serious issue that can reduce learning motivation, academic performance, and mental health. Approximately 25–30% of students experience burnout symptoms, which negatively affect concentration and academic productivity. Early detection is still limited due to the lack of accurate data analysis. This study aims to predict the risk level of student burnout using the C5.0 algorithm as a classification method capable of handling both categorical and numerical data. The research data were obtained from 306 students at Universitas Islam Negeri Sumatera Utara through an online questionnaire based on the Maslach Burnout Inventory–Student Survey (MBI-SS). The data were processed through cleaning, encoding, and splitting into training and testing sets using Python. The results show that the model achieves excellent classification performance, with an accuracy of 99.25% on the training set (precision 99.72%, recall 99.45%) and 97% on the testing set (precision 100%, recall 96%). The model also identifies the most influential attributes contributing to burnout, such as stress level and emotional exhaustion. The main contribution of this study is the development of an accurate and interpretable machine learning-based model for predicting student burnout risk. These findings provide practical implications for educational institutions in supporting early detection and designing data-driven preventive interventions, such as counseling services and stress management programs.]]>
