Article: An advanced method using machine learning algorithms to detect children with autism spectrum disorder Journal: International Journal of Computational Vision and Robotics (IJCVR) 2026 Vol.16 No.5 pp.1 - 17 Abstract: ASD, or autism spectrum disorder, is a brain disease which makes it hard for a person to learn words, talk to others, think critically, and connect with others throughout their whole life. According to the Autism Society, about 1% of people in the world have autism (https://www.autism-society.org/whatis/facts-and-statistics/). The first two years after birth, or the developmental periods, are when autism symptoms often become apparent, as of 25 December 2019. Even while genetics or environmental factors account for the majority of cases of ASD, early detection and treatment might potentially lessen symptoms. We prepared a dataset from the children who were affected by ASD. Thus, our study's main goal is to determine whether or not the child is susceptible to ASD in its early stages, to speed up the diagnosis procedure. Our results show that, for our dataset, logistic regression offers the highest degree of accuracy. Inderscience Publishers - linking academia, business and industry through research

Title: An advanced method using machine learning algorithms to detect children with autism spectrum disorder

Authors: Subair Ali Liayakath Ali Khan; K. Saravanan

Addresses: Department of Information Technology, PRIST University, Thanjavur, Tamil Nadu, India ' Department of Computer Science, PRIST University, Thanjavur, Tamil Nadu, India

Abstract: ASD, or autism spectrum disorder, is a brain disease which makes it hard for a person to learn words, talk to others, think critically, and connect with others throughout their whole life. According to the Autism Society, about 1% of people in the world have autism (https://www.autism-society.org/whatis/facts-and-statistics/). The first two years after birth, or the developmental periods, are when autism symptoms often become apparent, as of 25 December 2019. Even while genetics or environmental factors account for the majority of cases of ASD, early detection and treatment might potentially lessen symptoms. We prepared a dataset from the children who were affected by ASD. Thus, our study's main goal is to determine whether or not the child is susceptible to ASD in its early stages, to speed up the diagnosis procedure. Our results show that, for our dataset, logistic regression offers the highest degree of accuracy.

Keywords: autism spectrum disorder; ASD; machine learning; dataset; preprocessing; encoding; support vector machine; SVM; k-nearest neighbours; KNN; random forest; logistic regression; confusion matrix; precision; recall; F1 score; accuracy.

DOI: 10.1504/IJCVR.2026.153922

International Journal of Computational Vision and Robotics, 2026 Vol.16 No.5, pp.1 - 17

Received: 01 Jul 2024
Accepted: 03 Dec 2024
Published online: 08 Jun 2026 *

Title: An advanced method using machine learning algorithms to detect children with autism spectrum disorder

Keep up-to-date