To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Age-based Data Analysis and Prediction of Autism Spectrum Disorder |
|
Author |
Sujatha R, Aarthy SL, Jyotir Moy Chatterjee, NZ Jhanjhi, Azween Abdulla, Mahadevan Supramaniam |
| Citation |
Vol. 25 No. 4 pp. 95-104 |
|
Abstract |
Autism spectrum disorder (ASD) is a syndrome prevalent in all age groups that causes immense changes in all aspects of an affected person’s life, including social skills, communication, and behavioral style. Screening of the same is a challenging task, and classification must be conducted with great care. The dataset considered for this work is the benchmark dataset retrieved from the UC Irvine (UCI) Machine Learning Repository. The case sample considered here includes approximately 1,000 children of various autism spectrum conditions and age groups mapped as a child, adult, or adolescent. The autism or no autism class categorized based on the following attributes assessed include age, gender, ethnicity, born with jaundice, pervasive developmental disorder of any family member, information about a relationship who is undergoing the test, country of residence, screening methods. Autism spectrum quotients (AQs) varied among a number of scenarios for toddlers, adults, adolescents, and children that include positive predictive value for the scaling purpose. AQ questions referred to topics pertaining to attention to detail, attention switching, communication, imagination, and social skills. The diagnostic decision support system with the provided features for the ASD was optimized on the basis of the selected dataset with the help of machine learning algorithms and soft computing techniques. The dataset was classified by using various algorithms, and accuracies in the range of 85%?95% were obtained. |
|
Keywords |
Autism; International Classification of Disease; machine learning; soft computing; training set; testing set; cross validation; specificity; sensitivity; accuracy. |
|
URL |
