A Survey on Autism Spectrum Disorder (ASD) using Machine Learning

Abstract

At present, Autism is a disorder procuring at a drastic rate without any proper set of measures in order to halt it. Identification of characteristics of autism disorder through various types of tests is very expensive and diagnosis of it can be time-consuming in most of the cases. As the evolution of artificial intelligence and machine learning has created humongous impacts across various domains in our daily lives, it can be used to prognosticate autism with very little amount of time. Nevertheless, several types of research were conducted across different set of systems, the observations from these haven’t generated an efficient or a definite conclusion about forecasting autistic characteristics concerning the different levels of age groups. Hence, there is a need of technology or a type of system that deals with implementing an efficient and low-cost ML model which enhances the efficiency by giving accurate results. We also need to make sure that the system will also incorporate a large number of datasets accumulated from different set of individuals who may or may not possess autistic traits which is done outside of laboratory conditions “ethically”, for different groups of people.