Heart Disease Prediction Using Hybrid Random Forest and Linear Model

Abstract

Heart disease remains a significant global health concern, and accurate heart disease risk prediction plays a crucial role in its prevention. In recent years, machine learning techniques have shown promising results in cardiovascular disease prediction. This paper proposes a novel Hybrid Random Forest and Linear Model (HRFLM) that leverages diverse patient attributes to accurately predict the risk of developing heart disease. HRFLM utilizes a combination of feature engineering, feature selection, and an ensemble of machine learning algorithms to effectively capture the complex relationships between patient characteristics and heart disease outcomes. The model is trained and validated on a comprehensive dataset from the Framingham Heart Study, which includes a wide range of demographic, clinical, and lifestyle variables. The experimental results demonstrate that HRFLM
outperforms several state-of-the-art machine learning models like K-nearest neighbour (KNN), support vector machine (SVM), decision tree (DT), multilayer perceptron (MLP) and random forest (RF) in terms of accuracy, precision, recall, and F1 score. The proposed model provides valuable insights into the risk factors associated with heart disease and can assist doctors in identifying individuals at high risk for preventive interventions. The proposed model has achieved an accuracy of 88%.