Autism Prediction Using Machine Learning a Comparative Analysis of SVM And RNN Approaches

A neurological ailment ASD, also referred to as autism spectrum disorder, has a lifelong impact on an individual's capacity for communication and social interaction. Although it usually manifests itself in the first two years of life, it is categorised as a behavioral disorder and can be identified at any stage of life. According to the ASD idea, the difficulties start in childhood and continue until adolescence and maturity. Motivated by the increasing application of machine learning to medical diagnosis research, this study investigated the potential of random forest and support vector machines to predict and evaluate ASD-related problems in children. The suggested method is tested using non-clinical ASD datasets that are accessible to the public. Our proposed model achieves an overall accuracy of 97% for predicting autism in children.