A Hybrid Model for Predicting Cardiovascular Disease Based on Conventional Machine Learning Classification Algorithms

Abstract

In recent decades, heart disease, also known as cardiovascular disease, has been the leading cause of mortality Heart disease, sometimes called cardiovascular disease, has emerged as the major cause of death during the past few decades. Many different heart conditions fall under its umbrella. While a correct diagnosis of heart disease might lessen the likelihood of serious health issues, an incorrect one can be fatal. It incorporates a variety of cardiovascular disease risk factors with the need for time to acquire accurate, reliable, and reasonable techniques for early identification and speedy illness management. The World Health Organization (WHO) has concluded that cardiovascular disease is the biggest cause of death around the globe. An estimated 17.9 million people worldwide perish each year. Machine learning is widely used in healthcare as a data processing approach. For the purpose of better heart disease prediction, researchers investigate complex medical data using a variety of ML methods. In this paper, we present a number of variables associated with heart disease and a model built using supervised learning techniques such as Naive Bayes (NB), Decision Trees (DT), K-nearest Neighbor (K-NN), Logistic Regression (LR), the Random Forest (RF) algorithm, and Support Vector Machines (SVM). The Cleveland Cardiac Registry Dataset from the University of California, Irvine Heart Disease Patient Repository is used. There are 303 occurrences and 76 characteristics in the set. Only 14 of the 76 attributes listed above are really taken into account while testing algorithms. The purpose of this paper is to determine a patient's risk of developing heart disease. From what we can see, the Decision Tree scores highest in terms of accuracy.