Support Vector Machine for Diagnosis Autism Spectrum Disorder in Toddlers Intan Nurma Yulita, Aditya Rizky Fadillah
Padjadjaran University
Abstract
Autism spectrum disorder (ASD) is a developmental disorder that affects communication and behavior. ASD can be seen in children since toddlers. The knowing toddler has ASD is very important. It can anticipate as early as possible to minimize the worsening of ASD. It can be done by early detection. The mechanism is recognizing patterns form existing data to build detection models. The system can be built using machine learning mechanism. It has been widely used to help diagnose medical data. Support Vector Machine (SVM) is a method in machine learning. This study implemented it in the data which came from 1054 ASD patients. The study also compared the method to other machine learning methods. The results were obtained that SVM accuracy is higher than others. The SVM obtained an accuracy of 99.90%. It shows that SVM promising tool for the diagnosis of ASD.
Keywords: Autism Spectrum disorder, Machine Learning, Support Vector Machine, Toddler
