Date of Award
Spring 2016
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Mathematics, Statistics and Computer Science
First Advisor
Ahamed, Sheikh I.
Second Advisor
Franco, Zeno
Third Advisor
Praveen, Madiraju
Abstract
Many veterans undergo challenges when reintegrating into civilian society. These challenges include readapting to their communities and families. During the reintegration process veterans have difficulties finding employment, education or resources that aid veteran health. Research suggests that these challenges often result in veterans encountering serious mental illness. Post-Traumatic Stress Disorder (PTSD) is a common mental disease that veterans often develop. This disease impacts between 15-20% of veterans. PTSD increases the likelihood of veterans engaging in high risk behaviors which may consist of impulsivity, substance abuse, and angry outbursts. These behaviors raise the veterans’ risk of becoming violent and lashing out at others around them. In more recent studies the VA has started to define PTSD by its association to specific high risk behaviors rather than defining PTSD based on a combination of psychiatric symptoms. Some researchers have suggested that high risk behaviors -- extreme anger (i.e., rage or angry outbursts) is particularly problematic within the context of military PTSD. Comparatively little research has been done linking sensor based systems to identify these angry episodes in the daily lives of military veterans or others with similar issues. This thesis presents a middleware solution for systems that work to detect, and with additional work possibly prevent, angry outbursts (also described in psychological literature as “rage”) using physiological sensor data and context-aware technology. This paper will cover a range of topics from methods for collecting system requirements for a subject group to the development of a social-context aware middleware. In doing such, the goal is to present a system that can be constructed and used in an in lab environment to further the research of building real-world systems that predict crisis events, setting the state for early intervention methods based on this approach.