Evaluating the Performance of Three Classification Methods in Diagnosis of Parkinson’s Disease

Accurate diagnosis of the Parkinson’s disease is a challenging task that involves many physical, psychological and neurological examinations. The examinations include investigating a number of signs and symptoms, reviewing the medical history and checking the nervous system conditions of a patient. Recently, researchers use voice disorders to diagnose Parkinson’s disease patients. They extract features of a recorded human voice and apply classification methods to diagnosis this disease. In this paper, we apply a Decision Tree, Naïve Bayes and Neural Network classification methods for the diagnosis of Parkinson’s disease. The aim of this paper is to resolve the problem by evaluating the performance of the three methods. The objectives of the paper are to (i) implement three classification methods independently on a Parkinson’s dataset, and (ii) determine the best method among the three. The classification results show that the Decision Tree produces the highest accuracy rate of 91.63%, followed by the Neural Network, 91.01% and the Naïve Bayes produces the lowest accuracy rate of 89.46%. The results recommend using the Decision Tree or the Neural Network over the Naïve Bayes for datasets with similar properties.