Intelligent Heart Disease Prediction using Novel Random Forest Algorithm with Enhanced Accuracy over SVM

Abstract

The major goal is to use Random Forest Algorithm to develop a unique intelligent model for Heart Disease Prediction that is more accurate than Support Vector Machine. Materials and Procedures: Two machine learning techniques, the Random Forest algorithm (N=42) and the Support Vector Machine algorithm (N=42), are used to predict heart disease. Heart dataset, which is gathered from kaggle.com, is used to predict heart disease. The dataset is made up of 303 rows and 14 key heart-related characteristics. Twenty samples are gathered for each group, which are then split into training and testing groups. Results: Random Forest's accuracy is 94%, whereas SVM's accuracy is 67%. Random Forest and SVM algorithms have analytically significant differences with p0.05. Conclusion: heart disease utilizing prediction.