Predicting autism spectrum disorder using machine learning algorithms with jaundice symptomatic analysis

Abstract

We utilized a dataset identified with autism screening all age set of autism: toddler, child, adolescent, adult contained 20 attributes which are used for investigation particularly in deciding persuasive autistic traits, enhancing the order of ASD cases. With 10 social features in addition to 10 individual qualities that have ended up being successful in identifying the ASD cases, consequently applied RT to get the best clusters, process them through RF to get exactness. Primary objective of this work is to predict the correlation between the ASD with its symptoms by applying the machine learning techniques of the data science. The prescribed work is done to predict the correlation between the jaundice symptomatic patients, further progression of the same to ASD. This work also compares the chances of genetic influence which is the secondary classifier that leads to the disorder. To accomplish this objective, we applied our validated supervised Machine Learning, random tree, and random forest.