PhD Dissertation Defense - Hillol Sarker
From Markers to Interventions - The Case of Just-in-Time Stress Intervention
Hillol Sarker, PhD Candidate
Friday, November 4, 2016, 8:30 am
Dunn Hall 311
Wearable wireless sensors for health monitoring are enabling the design and delivery of just-in-time interventions (JITI). Critical to the success of JITI is to time its delivery so that the user is available to be engaged. This dissertation takes a first step in modeling users' availability by analyzing 2,064 hours of physiological sensor data and 2,717 self-reports collected from 30 participants in a week-long field study. Delay in responding to a prompt is used to objectively measure availability. Presented work compute 99 features and identify 30 as most discriminating to train a machine learning model for predicting availability. Findings suggest that location, affect, activity type, stress, time, and day of the week, play significant roles in predicting availability. Users are least available at work and during driving, and most available when walking outside. Proposed model finally achieves an accuracy of 74.7% in 10-fold cross-validation and 77.9% with leave-one-subject-out.
Management of daily stress can be greatly improved by delivering sensor-triggered just-in-time interventions (JITIs) on mobile devices. In addition to assessing the availability of a person, the success of such JITIs critically depends on being able to mine the time series of noisy sensor data to find the most opportune moments. This dissertation proposes a time series pattern mining method to detect significant stress episodes in a time series of discontinuous and rapidly varying stress data. This model is applied to two separate human subject studies on physiological, GPS, and activity data collected from 91 (38+53) users in their natural environment to discover patterns of stress in real life. Findings suggest that the duration and the type of a prior stress episode predict the duration and the type of the next stress episode. Stress in mornings and evenings is lower than during the day. The work then analyzes the relationship between stress and objectively rated disorder in the surrounding neighborhood and suggest a model to predict stressful episodes.