Improved Random Forest Classifier for Predicting Heart Disease

In recent years, heart disease, also known as coronary artery disease, has been the leading cause of death worldwide. It includes numerous heart-related conditions. In order to address a variety of risk factors for heart disease, it is essential to have prompt access to reliable, practicable, and accurate disease management and early detection techniques. In the manufacturing of health maintenance products, data mining is frequently used to manage enormous quantities of data. In order to forecast cardiac disease, researchers examine enormous quantities of complex medical data using a number of data extraction and machine-learning techniques. This project’s model relies on supervised learning techniques, such as DT, Random Forest (RF), K-Nearest Neighbour, and SVM Classifier. Numerous cardiac disease-related characteristics are prevalent. It uses the contemporary dataset from the Cleveland database derived from the UCI Coronary artery disease patient history. The collection contains 1026 instances with 76 characteristics. Only 14 of the 76 qualities that were essential for demonstrating the performance of diverse algorithms are utilised in the testing procedure. The purpose of this investigation is to predict the likelihood that a patient will develop cardiac disease. The Random Forest Classifier (RFC) Algorithm has the maximum accuracy rating, according to the results.