An Effective Machine Learning Model to Detect and Analyze the Autism Spectrum Disorder

Autism spectrum disorder is a neuro-developmental disability that impacts a person's interactions with other people. The primary indicators of this disease are loss of eye contact and an inability to react when called. Early diagnosis of this disorder is essential, and many machine learning algorithms are available to analyze and identify these disabilities for medical diagnosis. These models automatically identify the patient's self-stimulatory activities. The main aim of this proposed work is to analyze various Machine Learning algorithms such as Support Vector Machine, Adaboost, and Random Forest to forecast the possibilities of this disability and also identify the impact of this. For this purpose, all the possible algorithms are considered for evaluation and tested concerning various parameters, and the results are tabulated. From the analysis, it is observed that the Random Forest algorithm performs better than the other algorithms. The proposed analysis model produces around 14% better reliability than all other previously used models for the same input.