Klasifikasi tingkat depresi mahasiswa menggunakan algoritma decision tree

Abstract

Depression is a mental health disorder commonly experienced by college students due to academic, social, and financial pressures. The high rate of depression among college students can impact academic achievement, social relationships, and overall quality of life. Therefore, a method is needed to quickly and accurately identify levels of depression. This study aims to classify college students' depression levels using the Decision Tree algorithm. The method used includes utilizing the Student Depression Dataset obtained from the Kaggle platform, data preprocessing, building a classification model using the Decision Tree algorithm, and evaluating model performance. The Decision Tree algorithm was chosen because it produces decision rules that are easy to understand and interpret. The test results obtained an accuracy value of 81.08%. This value indicates that the model is capable of classifying college students' depression levels quite well. Furthermore, the classification report results shows that the model has a precision value of 0.81 for both classes. The recall value for the depression class reached 0.88, indicating that the model successfully recognized most students experiencing depression. The research findings can assist educational institutions in early detection of students at risk of depression, allowing for more effective treatment and support.