Improving the Efficiency of Heart Disease Prediction Using SVM and a Novel Tree Specific Random Forest Classifier (NTSRF)

Abstract

Aim: The objective of the work is to evaluate the accuracy and precision in predicting the heart disease using Support Vector Machine (SVM) and Random Forest (RF) classification algorithms. Materials and Methods: Random Forest Classifier is applied on a Health dataset that consists of 304 records. A framework for heart disease prediction in the medical sector comparing Random Forest and SVM classifiers has been proposed and developed. The sample size was measured as 21 per group. The accuracy and the precision of the classifiers was evaluated and recorded. Results: The SVM classifier produces 53.04% in predicting the heart disease on the data set used whereas the Random forest classifier predicts the same at the rate of 83.2%. The significant value is 0.0. Hence RF is better than SVM. Conclusion: The performance of Random forest is better compared with SVM in terms of both precision and accuracy.