Analisis Pola Tidur Dalam Produktivitas Belajar Menggunakan Algoritma Decision Tree Dan Random Forest Pada Mahasiswa Universitas Labuhanbatu

Abstract

Sleep patterns are one of the important factors that support cognitive function and students' learning productivity. Irregular sleep habits, insufficient sleep duration, poor sleep quality, and excessive use of gadgets at night can affect students' learning productivity. This study aims to analyze the influence of sleep patterns on the learning productivity of students at Universitas Labuhanbatu using the Decision Tree and Random Forest algorithms, as well as to compare the performance of both algorithms in classification tasks. The variables used in this study consist of Sleep Duration (X1), Sleep Quality (X2), Sleep Time (X3), Sleep Consistency (X4), and Nighttime Gadget Usage (X5) as independent variables, while Learning Productivity (Y) serves as the dependent variable. The research data were collected through questionnaires distributed to 80 students, which were then divided into 50 training data and 30 testing data. Data processing was carried out using the Orange Data Mining application through the stages of data preprocessing, model construction, and model evaluation using the Test and Score method with the evaluation metrics of Area Under Curve (AUC), Classification Accuracy (CA), F1-Score, Precision, Recall, and Matthews Correlation Coefficient (MCC). The results showed that the Decision Tree algorithm achieved an AUC of 0.931, CA of 0.920, F1-Score of 0.921, Precision of 0.925, Recall of 0.920, and MCC of 0.834. Meanwhile, the Random Forest algorithm achieved an AUC of 0.986, CA of 0.920, F1-Score of 0.921, Precision of 0.925, Recall of 0.920, and MCC of 0.834. Based on these results, the Random Forest algorithm demonstrated better performance in terms of the AUC value, making it more suitable for classifying students' learning productivity based on sleep patterns.