Study on Predicting Heart Disease Diagnosis with Hybrid Machine Learning Techniques

Abstract

Machine learning can successfully forecast cardiac disease. The main benefit of these systems is their adaptability in non-linear contexts, allowing them to handle new data sets. Heart illness is the most common. We examined many indicators to better predict heart illnesses and also applied algorithms to forecast them. Modernity encourages us to be more active and fit, but it also pushes us to push ourselves harder and risk injury. These ecosystem-wide advancements have given bacteria, viruses, and other diseases a substantial new capability in this setting. Heart failure seems to be on the rise. Blood pressure, sugar, heart rate, and other markers are cardiovascular risk factors that cause blood arteries to be restricted or locked. Aneurysm, heart, or stroke. It can cause heart disease, vascular disease, CVA, cardiac death, and sudden death. Medical exams are used to diagnose various cardiac conditions, but the patient's family history and other factors should be considered. It's more tough to conclude for folks who don't get checked and have heart failure. Heart disease is one of the most common ailments nowadays, and early detection is critical to saving lives. The goal of this article is to improve accuracy, reduce training time, and reduce unknown cases by evaluating multiple classifiers on the data set to discover optimal HD attribute configurations. The K-Nearest Neighbor (K-NN), Naive Bayes, and SVM were compared to represent, JR and Adrost Decision Tree (JRandom), in order to assess the potential