Analysis of Autism Spectrum Disorder in Children with Data Mining Methods

Data mining techniques aim to reveal hidden patterns in data. They are widely used in many fields, such as medicine. Autism spectrum disorder, whose diagnosis and treatment are difficult and lengthy, is a complex neurodevelopmental disorder that is congenital or occurs in the first years of life. Ac...

Full description

Saved in:
Bibliographic Details
Main Authors: Sümeyye Çelik, Melike Şişeci Çeşmeli
Format: Article
Language:English
Published: Istanbul University Press 2021-06-01
Series:Acta Infologica
Subjects:
Online Access:https://cdn.istanbul.edu.tr/file/JTA6CLJ8T5/4C99D0EF27B54DCC9135FF2A2219B25A
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data mining techniques aim to reveal hidden patterns in data. They are widely used in many fields, such as medicine. Autism spectrum disorder, whose diagnosis and treatment are difficult and lengthy, is a complex neurodevelopmental disorder that is congenital or occurs in the first years of life. Actual and current autism spectrum disorder data collected from 292 children were used in this study. The data set has 20 input attributes and 1 output attribute. The output attribute expresses whether autism is present or not. In the study, data preprocessing stages, such as completing missing data on the data set, digitizing categorical data, and normalization, were first carried out. Subsequently, the features were classified by artificial neural networks and linguistic strength neuro-fuzzy classifier and clustered with k-means and x-means. The results of each method were evaluated and the performances were compared.
ISSN:2602-3563