Intelligent decision tree-based software was built to aid in predicting the adolescent with autism traits. This application, which is obtained and operated on mobile devices, uses artificial intelligence and machine learning techniques to assign probabilities to people who pass the test in the application. The app has a knowledge base that assists in the prediction of autism in adolescents. This paper intends to demonstrate the feasibility of using fuzzy neural networks to assist in predicting the identification of autism traits, mainly supported by a system capable of generating fuzzy rules more cohesive than a decision tree. Therefore, this article proposes the insertion of an interpretive technique based on an extreme learning machine to deal with questions provided by users that seek to obtain more immediate answers, based on classification binary labels. The tests performed with the base achieved high levels of precision for the proposed model and support, making it a viable alternative for the efficient prediction of adolescents with autism.

Using Fuzzy Neural Networks Regularized to Support Software for Predicting Autism in Adolescents on Mobile Devices

de Campos Souza, Paulo Vitor;
2019-01-01

Abstract

Intelligent decision tree-based software was built to aid in predicting the adolescent with autism traits. This application, which is obtained and operated on mobile devices, uses artificial intelligence and machine learning techniques to assign probabilities to people who pass the test in the application. The app has a knowledge base that assists in the prediction of autism in adolescents. This paper intends to demonstrate the feasibility of using fuzzy neural networks to assist in predicting the identification of autism traits, mainly supported by a system capable of generating fuzzy rules more cohesive than a decision tree. Therefore, this article proposes the insertion of an interpretive technique based on an extreme learning machine to deal with questions provided by users that seek to obtain more immediate answers, based on classification binary labels. The tests performed with the base achieved high levels of precision for the proposed model and support, making it a viable alternative for the efficient prediction of adolescents with autism.
2019
9789811386138
9789811386145
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11582/341015
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact