A robust technique for identification of autism spectrum disorder using ensemble voting classifier

Abstract

Autism Spectrum Disorder, also known as ASD, is described as a disorder of communication and behavior. A person can receive a diagnosis at any time during their lifetime. Without regard to factors such as ethnicity, race, or economic status, the first two years of life are of the utmost importance. Depending on the severity and nature of the symptoms that individuals experience, there are a few distinct subtypes of ASD. Even though it is a disorder that lasts a lifetime, medical care and treatment can help alleviate the symptoms. A wide variety of research and clinical studies have been examined; however, only a few of them have provided satisfactory medical evidence for the strong distinction of ASD from healthy individuals. This paper presents a robust technique that employs ensemble voting classifier for identifying individuals who display specific symptoms of ASD. In addition, the purpose of this paper is to expedite the process of diagnosing autism in order to provide the necessary treatment at an earlier stage in the development of any person. The results of our experiment show that the Voting Classifier is more effective, with a higher identification accuracy of 99.9%.