Abstract
Autism Spectrum Disorder (ASD) is a developmental disability caused by differences in the brain. People with ASD often have problems with social communication and interaction, and restricted or repetitive behaviors or interests. People with ASD may also have different ways of learning, moving, or paying attention. In this project, we present a novel approach for the detection of ASD using machine learning techniques, implemented in Python. We employed two distinct algorithms, namely the Random Forest Classifier and the Decision Tree Classifier, to analyze a dataset containing some records with 21 features. The dataset includes a diverse range of attributes, such as sensory perception, cognitive abilities, demographics, and medical history, which are potentially indicative of ASD. Our model's performance on this dataset is a testament to the power of machine learning in healthcare applications.The Random Forest Classifier achieved remarkable results with a training accuracy score of 100% and a testing accuracy score of 99%. This indicates that the model can effectively learn from the training data and generalize well to unseen cases. The Decision Tree Classifier, while achieving a training accuracy score of 100%, maintained a testing accuracy score of 96%, showcasing robust performance. The dataset used in this project encompasses a comprehensive set of attributes, including sensory perception (A1_Score - A6_Score), cognitive abilities (A7_Score - A10_Score), age, gender, ethnicity, parental medical history (jundice), autism diagnosis (austim), country of residence, prior app usage, and various demographic features. The extensive range of attributes ensures that our model takes into account a multitude of factors when making predictions.