Improved Accuracy in Heart Disease Prediction using Novel Random Forest Algorithm in Comparison with Support Vector Machine Algorithm

Abstract

Aim: The primary aim of this paper is to improve the exactness of the classification of coronary illness predictions using the help of machine learning algorithms. We will be comparing the novel Random Forest with the Support Vector Machine to find out which of these can give us the best accuracy. Material and Methods: The study used 184 samples with Novel Random Forest and Support vector machine is executed with varying training and testing splits for anticipating the accuracy for coronary illness prediction with the 80% of G-power value and the Coronary disease data were gathered from several web sources and included current research findings, a threshold of 0.05%, a confidence range of 95%, a mean, and a standard deviation. The accuracy rate of the classifiers is used to evaluate their performance using the Coronary disease dataset. The difference between the two groups is statistically negligible. (p = 0.225; p > 0.05). Results and Discussion: The accuracy of predicting heart disease in Novel Random Forest is 90.16% and support vector machine is 81.97%. Conclusion: Expectation of heart infection using the Novel Random Forest (RF) algorithm appears to be significantly better than the Support Vector Machine (SVM) with improved accuracy.