An Analysis of Parkinson Disease Prediction Using Machine Learning Approaches

Abstract

Objective: The main goal of the study isto inspect the performance of three Supervised algorithms for improving the Parkinson disease diagnosis by detection. Methods: We used three machine learning techniques for the detection of Parkinson disease datasets. SVM, KNN, and LR were used for prediction of Parkinson Disease. The performance of the classifiers was evaluated via recall, precision, f 1 measure and accuracy. Results: SVM shows the accuracy level 100% for Parkinson disease prediction. LR achieved the second highest classification accuracy of 97%. Moreover, in the terms of accuracy for analyzing Parkinson disease datasets, KNN achieved the worst performance (i.e. 60%). Conclusion: Our finding showed that the SVM obtained the highest performance for analyzing the Parkinson datasets. This study has emphasized the current Parkinson research trends and scope in relation to clinical research fields by machine learning techniques. That will be an effective impact in the field of Parkinson disease.