Supervised Learning Approaches for Heart Disease Prediction: A Comparative Review

Abstract

As heart disease is among the causes of death in the world, there is need to have accurate and quick prediction tools to assist in clinical decisions. The research paper makes comparisons of numerous machine learning methods that apply regression and classification to predict cardiac disease. Although the regression models such as Linear Regression, Support Vector Regression (SVR), Decision Tree regression, and Random Forest were used to predict the risk scores, classification models such as Logistic Regression, SVM, Decision Tree, and Random Forest to identify the presence of the disease. Depending on the findings of the experiments, the ensemble forms of Random Forest models outperformed the other forms with regard to accuracy, R2 score and Mean Squared Error (MSE). The model different performance was also statistically significant. The proposed framework as a possible solution to collaborate with early diagnosis and risk evaluation in the medical fields offers a valid and efficient predicting system that balances in terms of accuracy and the complexity of the computation.