Early autism diagnosis of children with machine learning algorithms

Abstract

Autism Spectrum Disorder (ASD) is a neuro-developmental disorder that has become one of the major health problems, and early diagnosis has a great deal of important in terms of controlling the disease. The increase in the number of autoimmune influenza and ASD cases in the world reveals an urgent need to develop easily applied and effective screening methods In this study, performance comparisons were made using three different classification methods, Naive Bayes, IBk (k-nearest neighbors), RBFN (radial basis function network), and Random Forest, on UCI 2017 Autistic Spectrum Disorder Screening Data for Children dataset. As a result of the experiment, Random Forest method has been shown to be more successful than Naive Bayes, IBk and RBFN methods. © 2018 IEEE.