Enhancing Early Diagnosis of Autism Spectrum Disorder in Children: A Comparative Analysis of Machine Learning Techniques

Autism Spectrum Disorder (ASD), a mental condition characterized by impaired communication and repetitive behaviors, requires early detection in toddler children. Developing reliable prediction models depends critically on using relevant characteristics, while machine learning approaches have shown promise as tools for ASD identification. Because of their capacity to evaluate big datasets and derive insightful conclusions, machine learning algorithms have drawn much attention in ASD diagnosis. In this paper, multiple machine learning techniques for ASD identification are compared. Data has been collected from the Kaggle database containing behavioral and clinical information from people with ASD, and neurotypical controls are compiled with pre-generated data. Various machine learning methods are utilized, including logistic regression, support vector classification, naive Bayes, random forest, and k-nearest neighbor. The effectiveness of the models is assessed using performance metrics such as accuracy, precision, recall, and F1-score.