Publication

Abstract : Heart disease continues to be the world's top cause of death, which emphasizes the need for precise and fast prediction technologies. Machine learning (ML) algorithms have developed into crucial tools for predicting cardiac disease by utilizing the vast quantity of medical data that is already available. This abstract describes the key components of heart disease prediction using ML, including data collection, data description, data preprocessing, model development, and evaluation procedures. To increase the effectiveness and efficiency of the prediction models, pertinent features are selected from the dataset or extracted. The preprocessed data can be subjected to a variety of machine learning (ML) approaches, including decision trees, random forests, support vector machines, k-nearest neighbors, additional tree classifiers, logistic regressions, and support vector machines. Finally, fitting analysis and prediction accuracy were investigated using the top-ranked ML classifier. Extra Tree (ET) Classifiers have been found to be the most effective at predicting heart disease, with testing score of 0.9761285935, training score of 0.997936175, and accuracy values of 0.9720963844. The incorporation of domain knowledge and medical experience within the ML framework makes the predictions more understandable and beneficial for clinical decision-making. The suggested strategy for predicting cardiac disease shows that powerful self-sustaining diagnostic systems are possible. Finally, the use of machine learning to forecast cardiac illness offers hope for accuracy and early detection. As machine learning advances, it has the potential to revolutionize the prediction of cardiac disease and significantly improve patient care and outcomes.