Title: Machine learning techniques for autism spectrum disorder (ASD) detection

Authors: Anshu Sharma; Poonam Tanwar

Addresses: Department of Computer Science, Manav Rachna International Institute of Research and Studies, Faridabad, 121006, Haryana ' Department of Computer Science, Manav Rachna International Institute of Research and Studies, Faridabad, 121006, Haryana

Abstract: Autism spectrum disorder (ASD) which is termed as ASD is a compound, integrated and lifelong growing incapability which comprises problem that are distinguished by repetition in behaviour, communication (non-verbal), doziness. In recent years, Autism is growing at a massive momentum which needs timely and early diagnosis. Autism can be detected through various tools (screening), but it is very time consuming and costly. In past few year, for prediction of ASD different types of dataset are used like images of autistic and non-autistic children, behavioural feature, genetic dataset etc. These datasets can be processed on different mathematical model's life machine learning, recognition of patterns and so on. The main aim of this paper is to analyse different types of datasets used to predict the autism traits in children by various researcher with the help of techniques like support vector machine (SVM), random forest scan, decision trees, logistic regression etc. and contrast the result in terms of their efficiency and accuracy.

Keywords: machine learning; deep learning; classifier; genetic; dataset; behavioural features.

DOI: 10.1504/IJFE.2021.118912

International Journal of Forensic Engineering, 2021 Vol.5 No.2, pp.111 - 125

Received: 16 Jul 2020
Accepted: 05 Jan 2021
Published online: 11 Nov 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article