A COMPARATIVE STUDY OF MACHINE LEARNING APPROACHES FOR CHRONIC KIDNEY DISEASE PREDICTION

Chronic Kidney Disease (CKD) poses a significant global health challenge, and early diagnosis is vital to mitigate complications and improve patient outcomes.The study explores six machine learning algorithms—Logistic Regression (LR), Decision Tree (DT), K-Nearest Neighbors (KNN), Gradient Boosting Decision Tree (GBDT), CatBoost, and LightGBM—using a dataset with 25 health-related features.

By evaluating models through metrics like Accuracy, ROC AUC, Precision, Recall, F1-Score, and Average Precision, the research demonstrates the superiority of ensemble methods such as GBDT, CatBoost, and LightGBM, which achieved over 98% accuracy and ROC AUC. The findings emphasize the effectiveness of predictive modeling for CKD diagnosis, providing valuable insights into algorithmic trade-offs and real-world applicability.This work is relevant to PR&AI’s focus on artificial intelligence, pattern recognition, and intelligent systems, contributing to the journal’s mission of advancing information science and technology.