Prediction of Parkinson’s Disease Using Machine Learning Algorithms Comparative Analysis

Abstract

Parkinson’s disease is one of the major brain disorders which causes several uncontrollable movements such as balance difficulties, unintended shaking, and stiffness among other several symptoms. Several machine learning algorithms have been proposed to predict Parkinson’s disease where different classifiers have been utilized. These classifiers belong to different learning strategies. In this paper, an investigation regarding the best classifier that can predict Parkinson’s disease is proposed. The investigation considered one dataset and several evaluation metrics such as Accuracy, TP ratio, FP ratio, and F1-measure. The results revealed that IBK classifier achieved the best predictive performance compared with the other twenty-three classifiers and considering the previously mentioned metrics.