ML4Beats: A Hypertuning-Based Approach Towards Enhancement in Accuracy of Heart Disease Prediction Using Machine Learning

Abstract

Heart disease remains one of the leading causes of mortality worldwide, necessitating the development of accurate and efficient diagnostic tools. In this study, we propose ML4Beats, a hypertuning-based machine learning approach designed to enhance the accuracy of heart disease prediction. Our methodology involves the application of various supervised learning algorithms—such as Random Forest, Support Vector Machine (SVM), Gradient Boosting, and K-Nearest Neighbors (KNN)—integrated with hyperparameter tuning techniques including Grid Search and Randomized Search. By systematically optimizing model parameters, we aim to reduce overfitting and improve generalization across diverse datasets. The system is trained and evaluated on publicly available heart disease datasets, where performance metrics such as accuracy, precision, recall, and F1-score are employed for evaluation. Experimental results demonstrate that hypertuning significantly boosts model performance, with the best-tuned model achieving notable improvements over baseline implementations. Additionally, feature importance analysis helps in identifying critical medical attributes influencing prediction accuracy. ML4Beats thus presents a reliable, data-driven framework that supports clinicians in early diagnosis and risk assessment, contributing to more informed healthcare decisions. The findings confirm that intelligent model tuning can play a pivotal role in enhancing the reliability of machine learning systems in critical medical domains.