<![CDATA[Chronic Kidney Disease (CKD) is a serious health problem with increasing prevalence worldwide. Early detection is important to prevent severe kidney failure and reduce healthcare costs. This study aims to implement and compare the performance of Decision Tree C4.5 and Naïve Bayes algorithms for CKD prediction using a dataset from the UCI Machine Learning Repository containing 400 patient records and 24 clinical attributes. Data preprocessing included handling missing values, transforming categorical data into numerical form, and selecting relevant attributes. Model evaluation was conducted using 10-Fold Cross Validation with performance indicators including accuracy, precision, recall, and Area Under Curve (AUC). The results show that Decision Tree C4.5 achieved 93.00% accuracy, 84.27% precision, 100% recall, and 0.944 AUC. Meanwhile, Naïve Bayes obtained slightly better performance with 93.50% accuracy, 85.23% precision, 100% recall, and 0.948 AUC. The findings indicate that both algorithms are effective for CKD classification, although Naïve Bayes demonstrated better predictive performance, while Decision Tree C4.5 provided more interpretable classification rules. This research contributes to the development of intelligent decision support systems for early CKD diagnosis.]]>
Copyrights © 2025
