Detection Of Parkinson Disease Using Multiclass Machine Learning Approach

This article discusses the use of K-Nearest Neighbor (KNN) and Feed-forward Neural Network (FNN) models to improve the classification of Parkinson’s disease based on voice signal characteristics. The models utilize advanced optimization strategies such as SMOTE, feature selection, and hyperparameter tuning. The study uses a publicly accessible dataset from the University of Oxford and the National Canter for Voice, which includes voice recordings from individuals with PD and healthy controls. The results show promising improvements in PD classification using the proposed methods.