Feature Weighting for Parkinson's Identification using Single Hidden Layer Neural Network

The diagnosis of Parkinson has become easier with the existence machine learning. It includes using existing features from biometric dataset generated by person to identify whether he or not. differ in their discrimination capability and they suffer redundancy. Hence, researchers have recommended feature selection for Parkinson's identification. aims at finding most important relevant produce an efficient effective model. In this article, we present entropy-based classification. goal is select only 50% prediction. Two variants neural networks are used evaluation, first one a feed-forward Extreme Learning Machine ELM second Fast FLN. Also, K-Nearest Neighbor KNN algorithm evaluation. results show superiority FLN when model accuracy 80% compared 78% not used.