Paper Title
SMOTE and Ensemble Learning for Early Chronic Kidney Disease Diagnosis

Abstract
Chronic Kidney Disease (CKD) is a significant global health issue, impacting millions of people and contributing to high rates of mortality and morbidity. Since CKD often progresses without noticeable symptoms, it is frequently diagnosed at advanced stages, highlighting the critical need for early detection. Traditional diagnostic methods, which rely on clinical evaluations and biochemical tests, may not be sufficiently sensitive for identifying CKD in its initial stages. To bridge this gap, our study introduces a machine learning framework designed to detect CKD at an earlier stage. Key predictive features are identified using univariate feature selection, and the Synthetic Minority Over-sampling Technique (SMOTE) is employed to handle class imbalances in the dataset. Several machine learning models, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression (LR), and Gradient Boosting, using clinical, demographic, and biochemical data were evaluated. The Gradient Boosting model demonstrates superior performance, achieving 95% accuracy, 0.90 precision, and 0.97 recall. These findings highlight the potential of machine learning in improving CKD detection, enabling earlier and more accurate diagnoses than conventional methods. This approach could enhance patient outcomes and alleviate the healthcare challenges posed by CKD. Keywords - Early Diagnosis, Chronic Kidney Disease(CKD), Data Imbalance, Synthetic Minority Over-sampling Technique (SMOTE), Feature Selection, Predictive Modeling,Ensemble Techniques, Healthcare Analytics.