<![CDATA[Chronic kidney disease (CKD) is a global health issue that requires accurate diagnosis to prevent errors and unwanted side effects. This study aims to develop a reliable classification model using the XGBoost algorithm and to explore the effectiveness of the SMOTE method in addressing data imbalance. The dataset is sourced from the UCI Machine Learning Repository, consisting of 400 patient records with 25 attributes. The results indicate that the developed model performs well, with evaluation metrics (Accuracy, Precision, Recall, F1-Score, and AUC-ROC) nearing 1.0. The research findings reveal that the model without SMOTE is slightly superior, achieving an accuracy of 0.9874 compared to 0.9811 for the model with SMOTE. Analysis shows that the data imbalance is not significant, and XGBoost is resilient to unbalanced data. This study also identifies key factors influencing CKD diagnosis, such as hemoglobin and albumin, which can assist medical professionals in making more accurate diagnoses.]]>
Copyrights © 2026
