Comparison of Machine Learning Algorithms for Autism Spectrum Disorder Classification

DOI

10.2991/aer.k.201124.028How to use a DOI?

Keywords

Autism Spectrum Disorder, KNN, SVM, Random Forest, Deep Learning

Abstract

Autism Spectrum Disorder is one of the fastest-growing neurodevelopmental disorders in the world. These neurodevelopmental disorders often attack children, affecting social development and behavior. Effective early detection of ASD is needed to reduce the risk of ASD in children. This study classified the ASD using KNN, SVM, Random Forest, Backpropagation, and Deep Learning. The Children ASD dataset consists of 292 subjects, 49 ASD subjects and 243 normal subjects. This adolescents ASD dataset consisting of 104 subjects with 63 ASD subjects and 41 normal subjects. The classification process is done by applying some parameters to each algorithm. The algorithms performance in classifying ASD dataset was compared. Based on the specificity and sensitivity value of the Random Forest algorithm with full features is the best algorithm compared to other algorithms in classifying ASD in children and adolescents.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Erina S. Dewi
AU  - Elly M. Imah
PY  - 2020
DA  - 2020/11/24
TI  - Comparison of Machine Learning Algorithms for Autism Spectrum Disorder Classification
BT  - Proceedings of the International Joint Conference on Science and Engineering (IJCSE 2020)
PB  - Atlantis Press
SP  - 152
EP  - 159
SN  - 2352-5401
UR  - https://doi.org/10.2991/aer.k.201124.028
DO  - 10.2991/aer.k.201124.028
ID  - Dewi2020
ER  -