Predictive Modeling for Early Detection of Heart Disorders

Abstract

Heart disease remains a leading cause of mortality worldwide, emphasizing the critical need for early detection tools. This project applies machine learning techniques to predict heart disease using clinical features from the UCI dataset. After preprocessing and feature engineering, several algorithms were evaluated including Logistic Regression, Decision Tree, Random Forest, Naive Bayes, and XGBoost. A hybrid model was also developed using ensemble voting to improve accuracy. Evaluation metrics such as accuracy, precision, recall, and F1-score were employed for model comparison. Results revealed that the ensemble model outperformed individual algorithms, demonstrating high reliability and diagnostic potential. This research showcases how data-driven systems can assist medical professionals in early cardiac risk identification and improve patient outcomes.