Autism spectrum disorder (ASD) is a neuro disorder in which a person's contact and connection with others has a lifetime impact. In all levels of development, autism can be diagnosed as a “behavioural condition,” since signs generally occur within the first two years of life. The ASD problem begins with puberty and goes on in adolescence and adulthood. In this chapter, an effort is being made to use the supporting vector machine (SVM) and the convolutionary neural network (CNN) for prediction and interpretation of children's ASD problems based on the increased use of machine learning methodology in the research dimension of medical diagnostics. On freely accessible autistic spectrum disorder screening dates in children's datasets, the suggested approaches are tested. Using different techniques of machine learning, the findings clearly conclude that CNN-based prediction models perform more precisely on the dataset for autistic spectrum disorders.
PERFORMANCE OF CONSTRUCTIVE MACHINE LEARNING TECHNIQUES TO IDENTIFY THE BEHAVIORAL TRAITS OF AUTISM SPECTRUM DISORDER CHILDREN
