Interpretable Fuzzy System for Early Detection Autism Spectrum Disorder

Autism spectrum disorder (ASD) is a chronic developmental impairment that impairs person's ability to communicate and connect with others. In people ASD, social contact reciprocal communication are continually jeopardized. People ASD may require varying degrees of psychological aid in order gain greater independence, or they ongoing supervision care. Early discovery results more time allocated individual rehabilitation. this study, we proposed the fuzzy classifier for classification tested its interpretability index Nauck's ensure reliability. Then, rule base created Gauje tool. The rules were then applied neural network predict autism. suggested model built on Mamdani set optimized using backpropagation algorithm. uses heuristic function pattern evolution classify dataset. evaluated benchmark metrics accuracy F-measure, employed quantify interpretability. superior accurately detect an average rate 91% compared other classifiers.