Classification Autistic Spectrum Disorder with Selected Features

Abstract

In this paper, feature selection technique (FST) namely Chi-Square (CS) has been used for feature selection. The filter based CS is a ranking method. The FST key goals of improving classification efficiency and reducing feature counts. Naive Bayes (NB), K-Nearest-Neighbour (K-NN) and Support Vector Machine (SVM) with RBF kernel considered the classification methods on Autistic Spectrum Disorder (ASD) children dataset. Comparison to the non-reduced features and reduced feature of ASD datasets the reduced feature give up enhanced results in all classifiers NB, K-NN and SVM. Finally, minimum feature with high accuracy based classification model is proposed.