Abstract:
Machine learning has revolutionized various industries worldwide, including healthcare, by enabling predictive analytics for disease diagnosis and prognosis. One of its most promising applications is in the early detection of heart disease, which allows medical professionals to tailor treatment strategies and improve patient outcomes. The goal of this research is to apply machine learning methods to forecast an individual's risk of developing heart disease.
To achieve this, we evaluate various classification models, including Random Forest, Naïve Bayes, Support Vector Machines (SVM), Logistic Regression, and Decision Trees. Furthermore, we propose an ensemble approach that integrates both weak and strong classifiers to enhance prediction accuracy. In pursuit of improved classification performance, we also explore advanced boosting techniques such as AdaBoost and XGBoost. By analyzing these models and their effectiveness using key performance metrics—including confusion and correlation matrices—our research aims to develop a robust predictive framework for heart disease detection.
