Speaker: Julian Wolfson, University of Minnesota
Abstract: Many Americans carry a smartphone with them for a large part of their day. As a result, phone-based passive sensors can yield detailed information about an individual's location, and the types of activities (e.g., at home, driving, working, shopping, walking, eating out) they are engaged in at those locations. While major tech companies have sought to monetize such data by delivering context-specific advertising, they are also potentially of great value for biomedical research, by providing quantitative measures of health-related time use and well-being while capturing context that is vital for making sense of wearable sensor and device-based measurements. In this talk, after briefly introducing researcher-facing tools that can be used to collect sequential human activity data in consented studies, I will describe a range of novel statistical techniques that can be used to tackle key tasks with these data including clustering, inference, and synthesis.
Julian will be on-site for the seminar in W2008, but the seminar will also be streamed through Zoom.
ZOOM
Meeting ID: 928 8613 3715
Passcode: 838290
