An Innovative Penalty based Heart Disease Prediction system using Novel Random Forest over Logistic Regression Classifier Algorithm

Abstract

Aim: The main goal of the research is see how accurately predicting heart disease by Logistic Regression (LR) and Novel Random Forest(RF) Classifications. Materials and Methods: Novel Random forest appealed on a heart dataset which consists of 200 records A framework for predicting heart disease in the medical field has been proposed and developed to compare the RF with a LR classifier. The sample size was calculated to be 55 for each group with 80% G performance. The sample size was calculated using a Clincalc analysis with Alpha and Beta values of 0.05 and 0.5, pretest performance of 80%, and enrollment rate of 1. The Accuracy of the classifier was Evaluated and Recorded. Results: The LR produces 89.0% in predicting the heart disease on the data set used whereas the Novel Random forest classifier predicts the same at the rate of 95.46% of the time with a statistically significant difference between the two groups (P=0.03; P<0.05) with confidence interval 95%. Conclusion: RF is better compared with LR in terms of both precision and accuracy.