Prediction of Autism Spectrum Disorder based on Machine Learning Approach

Abstract

Abstract: In recent years, awareness of autism spectrum disorder (ASD) has grown faster than before. As everyone is aware, ASD is a disorder of neurodevelopment that also encompasses problems with conduct and social interaction. The degree of symptom severity and each individual's experience with ASD vary. Any age can be used to diagnose autism. According to research, violence, self-harm, elopement. tantrums, preoccupation, and lack of obedience are behaviour patterns most frequently observed in people with autism. Therefore, it is imperative to spot any sign of severe ASD as soon as possible. ML methods like Random Forest, Naive Bayes, Decision Tree, K- Nearest Neighbour, Logistic Regression, and Support Vector Machine (SVM) are employed in this work. With the use of algorithms that have been used to decide which gives accurate results in terms of speed and accuracy, performance metrics assist in examining the degree of correctness of each piece of data across all users. ML methods like Random Forest, Naive Bayes, Decision Tree, K-Nearest Neighbour, Logistic Regression, and Support Vector Machine (SVM) are employed in this work. With the use of algorithms that have been used to decide which gives accurate results in terms of speed and accuracy, performance metrics assist in examining the degree of correctness of each piece of data across all users.