A Comparative Analysis of Different Feature Selection Methods on a Parkinson Data

This paper provides a comparative analysis of different feature selection methods in a voice Parkinson dataset in order to find an optimal subset with relevant features which gives the higher accuracy. Filter and wrapper methods, and Genetic algorithm are considered for selecting the features. The performance of each feature selection method is evaluated through the accuracy of three popular supervised learning algorithms. Generalized Simulated Annealing (GSA) in used in improving the accuracy of the hyper-parameters of the classifiers. In conclusion, comparisons between the above mentioned methods are presented and the effect that has GSA in the accuracy for each subset.