Higher Prediction Accuracy on Parkinson Disease Patients Using Random Forest Algorithm over DTA

Abstract

Improving the prediction accuracy for Parkinson's patients is achieved by applying Innovative Parkinson's disease prediction utilizing classifiers that use machine learning methods and evaluating their performance. In this proposed work, the Innovative Parkinson's disease prediction has been carried out using the Random Forest algorithm and the Decision Tree algorithm. It was tested over a dataset consisting of 757 records. Both algorithms were subjected to a programming experiment in which N=10 iterations were used to discover the symptoms of Innovative Parkinson's disease prediction and their accurate analysis. The G-power test is around 80% accurate. From the implemented experiment by performing the independent sample t-Test, the Random Forest algorithm's Parkinson's disease prediction accuracy is significantly (0.028) better than the Decision Tree algorithm. The accuracy of Innovative Parkinson's disease prediction was compared between two algorithms, and the Random Forest algorithm appears to be higher at 93% than the Decision Tree algorithm's accuracy of 91%. This research will use the most up-To-date Machine Learning Classifiers to create an innovative Parkinson's disease prediction technique for the early detection of Parkinson's disease and other related issues.