Prognosis Of Idiopathic Parkinsonism Using Support Vector Machine And Random Forest Classifiers

Abstract

Many of individuals all over the world suffer with Idiopathic Parkinsonism (IP) which is more common in people over 50. Even today, despite numerous technological developments and breakthroughs, early disease identification is still difficult. This calls for the development of machine learning-based automatic methods that assist doctors in accurately identifying this disease in its early stages. This research paper's main goal is to give a thorough analysis of and comparison of the current machine learning methods used for IP detection. In order to compare and determine which of the two classifiers is the most effective and accurate for classifying IP, this paper discusses Support Vector Machine and Random Forest on a dataset. Accuracy and Kappa scores for support vector machine is 85.6% and 0.814. Accuracy 86.45% and kappa score of 0.81 was found in random forest.