Predicting Heart Diseases Using Machine Learning and Different Data Classification Techniques

Abstract

The main cause of worldwide dying may be prevented in the first -class element in the first identity of its symptoms and symptoms.  Due to complicated clinical records and difficulty offering continuous monitoring, the exact prognosis of heart disease is still hard.  Significant predictors have been discovered by many strategies of features that include Anova F-statistical (Anova FS), Chi-Squared Test (Chi2 FS) and Mutual Information (Mi FS) Using Data Set of Heart Diseases.  The data imbalance has become solved and the overall performance of the version was improved using the Synthetic Minority (SMOTE) technique.  Several system studies of fashion and file strategies have been used in the radical classification strategy.  Among them, a stacking classifier involving reinforced decision -making trees, extra trees and LightGBM, which led to huge results that achieved 100% accuracy in all approaches to selecting functions.  Excessive performance, emphasizing the opportunity to mix robust selections with state -of -the -art category modes for accurate medical facts, illustrates the effectiveness of the superior set that has been recognized in generating stable forecasts of heart disease.  This strategy shows how it is able to support early prognosis and results with a higher affected person.