Date of Award
Fall 12-2021
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computational and Data Sciences
First Advisor
Erik Linstead
Second Advisor
Rene German
Third Advisor
Dennis Dixon
Abstract
Machine learning and deep learning methods are becoming increasingly used in the understanding, identification, and improvement of the diagnosis and treatment of Autism Spectrum Disorder. People with ASD often exemplify challenging behaviors that can put their safety, education, and general quality of life at risk. Challenging behaviors are driven by one of four functions. The combination of common occurrences of challenging behaviors and their respective behavioral functions are unique to the individual and circumstance, and the most successful therapies account for both challenging behaviors and their respective functions. Therefore, it is important that research is done on these concepts to lead to improvements in therapy and outcomes.
In this thesis, we apply a cluster analysis to a sample of 1,416 individuals with Autism Spectrum Disorder. The aim is to find groupings of patients based on the relative frequency of each unique challenging behavior and function pair. As the first machine learning study to focus on combining the behavioral functions and challenging behaviors of ASD, we find that there are some patterns to be found based on eight identified clusters. The results of the study could impact the way that treatment and therapy plans are paved for children with Autism Spectrum Disorder.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
E. Daskas "Identifying functional profiles of challenging behaviors in Autism Spectrum Disorder with unsupervised machine learning," M. S. thesis, Chapman University, Orange, CA, Year. https://doi.org/10.36837/chapman.000330
Included in
Analytical, Diagnostic and Therapeutic Techniques and Equipment Commons, Mental and Social Health Commons