<![CDATA[The neurological disorder known as autism spectrum disorder (ASD) has an impact on a person's behavior, social communication, and interest patterns. Both repeated habits and communicative skills are lacking in this developmental condition. The World Health Organization (WHO) estimates that one out of every 100 youngsters worldwide has ASD. The Indonesian Ministry of Health in 2021 showed data on the increasing number of children with autism, which reached around 2.4 million with cases reaching 500 children every year. The use of machine learning can help classify and predict ASD based on health parameters. Using the Naive Bayes algorithm and Diagnostic and Statistical Manual of Mental Disorders 5 (DSM-5) data, this study attempts to create a classification application for people with ASD and assess how the model performs in grouping people with ASD. The results showed that the classification model developed produced optimal performance with an accuracy value reaching 95% while the highest precision, recall and F1-score values reached 100%. Evaluation using the macro average metric resulted in a precision value of 94%, recall of 87%, and f1-score of 90%. The weighted average metric produces positive precision, recall, and F1-score values of 95%. The developed model is integrated into a web-based application that features real time early screening and storage of user prediction results. The development of this application is expected to facilitate early screening so as to help determine effective interventions for individuals with autism spectrum disorders, thus making a positive contribution to the treatment of this disorder in daily practice.]]>
