<![CDATA[Depression is one of the most common mental health disorders among university students and may adversely affect academic performance and social functioning. The severity of depression can be assessed using the Patient Health Questionnaire-9 (PHQ-9), which classifies individuals into several levels of depression severity. This study aims to compare several machine learning models for multiclass classification of student depression levels based on PHQ-9 scores. The study employed the PHQ-9 Student Depression Dataset consisting of 682 student records. Predictor variables included age, gender, the nine PHQ-9 items, sleep quality, study pressure, and financial pressure. The models evaluated were Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and XGBoost. Model performance was assessed using accuracy, precision, recall, and F1-score metrics. The results indicate that XGBoost achieved the best performance, with an accuracy of 78,10%, macro precision of 0,77, macro recall of 0,77, and macro F1-score of 0,77. These findings demonstrate that XGBoost provides relatively good performance in the multiclass classification of student depression levels. This study suggests that machine learning approaches have the potential to support the identification of depression severity among university students.]]>
Copyrights © 2026
