Date of Award
2025
Degree Type
Thesis
Degree Name
Master of Science in Statistics
Department
Computer Science and Statistics
First Advisor
Gavino Puggioni
Abstract
his study explores factors associated with autism spectrum disorder (ASD) through analysis of a rich dataset comprising maternal and child characteristics. The data were carefully cleaned and preprocessed to address missing values and ensure analytical consistency. Statistical analyses and advanced visualization techniques were used to uncover patterns linking ASD status with variables such as maternal age, prenatal smoking, and education level. To further assess predictive capacity and feature importance, multiple supervised machine learning algorithms - logistic regression, decision trees, random forests, and support vector machines (SVM) - were applied. ASD appears to be influenced by a complex interplay of demographic and environmental factors. This study underscores the importance of integrative, data-driven approaches in ASD research and highlights the potential of machine learning techniques to uncover complex patterns in developmental disorder etiology.
Recommended Citation
Biswas, Supti, "RISK FACTORS FOR AUTISM SPECTRUM DISORDER: A STATISTICAL AND MACHINE LEARNING PERSPECTIVE" (2025). Open Access Master's Theses. Paper 2625.
https://digitalcommons.uri.edu/theses/2625