Open AccessAccuracy, Recall, Precision of SVM Kernels in Predicting Autistic Spectrum Disorder In Adults
Author(s) -
Didik Setiyadi,
Muhammad Dwison Alizah*,
Yulius Paulus Dharsono,
Sabar Sautomo,
Sfenrianto Sfenrianto
Publication year - 2020
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.f7655.038620
Subject(s) - support vector machine , recall , autism spectrum disorder , precision and recall , artificial intelligence , machine learning , kernel (algebra) , computer science , sigmoid function , pattern recognition (psychology) , artificial neural network , psychology , developmental psychology , autism , mathematics , cognitive psychology , combinatorics
Autism is a disorder that is quite difficult to diagnose when the condition of the sufferer is in the adult category. In this era, technology has been able to make predictions including health cases. Autistic Spectrum Disorder (ASD) in adults is felt to be predictable by using machine learning. This study will build a predictor for ASD sufferers. Predictors of machine learning are built using the Support Vector Machine (SVM) algorithm, with the type of kernel used was Gaussian RBF, Polynomial and Sigmoid. From the predictors that are built, the best SVM parameters will be searched based on accuracy. This best parameter is used to build the best new predictor and the results of the prediction are compared in terms of accuracy, recall, and precision. These results can be used to get the best performance when detecting ASD sufferers effectively and efficiently