Abstract

Now a days, Heart Disease (HD) has become a significant worldwide challenge, creating a need for early detection and intervention. Machine learning (ML) techniques are used effectively for assisting, decision making, and prediction in the healthcare industry. The classification algorithms, have demonstrated promising capabilities in predicting the risk of HD based on patient data. This paper provides a detailed overview of various ML algorithms for early detection of the risk of HD and focus on the study of eight algorithms namely "Logistic Regression", "Naive Bayes", "Support Vector Machine", "K-Nearest Neighbors", "Decision Tree", "Random Forest", "XGBoost", and “Neural Network". It also discusses a comparative study of their performance measures. Additionally, it explores the integration of feature selection and ensemble methods to increase predictive accuracy and generalization. The paper actually worked on these eight algorithms and observed that the accuracy of Random Forest Model – 95.08% is found to be the best accuracy amongst all. This paper aims to assist the researchers for selecting appropriate approaches and methodologies for the studies related to HD. The paper concludes with suggestions for future research directions aimed at advancing the field of HD prediction using ML algorithms.