WCSE

Diagnostic of Autism Spectrum Disorder

Abstract— According to the World Health Organization, Autism Spectrum Disorder affects 1 in 160 children, a disturbance of neurodevelopment characterized by symptoms such as difficulties in interaction and social communication, narrow interests, and repetitive behaviors. In this work, we diagnose Autism Spectrum Disorder using Machine Learning (ML) tools through the supervised training of Multi-Layer Perceptron and KNN classifiers. The validation algorithms were Hold-Out and K-Fold Cross Validation for both methods. The precision with the Multi-Layer Perceptron was 100% with Hold-Out and K-Fold Cross Validation, and with the KNN classifier, the accuracy was 91% and 87%, respectively.

Index Terms—Random forest classifier, online learning platforms, TAM, delone and mclean IS success model.

Elena Acevedo
Instituto Politécnico Nacional. ESIME Zacatenco, Mexico
Dinora Orantes
Instituto Politécnico Nacional, Centro de Investigación en Computación, Mexico
Antonio Acevedo
Instituto Politécnico Nacional. ESIME Zacatenco, Mexico

[Download]

Cite: María Elena Acevedo, Sandra Dinora Orantes, Marco Antonio Acevedo, "Diagnostic of Autism Spectrum Disorder," WCSE 2022 Spring Event: 2022 9th International Conference on Industrial Engineering and Applications, pp. 253-260, Sanya, China, April 15-18, 2022.