Paper Contents
Abstract
Being able to predict heart disease early and accurately is a powerful way to save lives and reduce illness, as it allows doctors to step in sooner. In this research, we explore different machine learning (ML) techniques to see if they can predict heart disease using common patient information that hospitals already collect. We test a variety of models, from classic methods like Logistic Regression and Random Forest to newer ones like XGBoost and neural networks. We train and test these models on a well-known public dataset, measuring how well they perform using metrics like accuracy, precision, and recall. We also use special tools (SHAP and LIME) to figure out why a model makes a certain prediction. This helps us see which health factors are most important and makes the models easier for doctors to trust. Our findings suggest that powerful models like Random Forest perform very well.
Copyright
Copyright © 2025 Sakshi Tekale, Sakshi Lande, Prof. Nitin Ganeshar. This is an open access article distributed under the Creative Commons Attribution License.
