A mobile context-aware behavior modification system for healthy lifestyle management

There are many online health-management services that send periodic motivational messages for exercising and eating right, but they lack any connection to a person's current physical location or state, increasing the chances of the message's advice being ignored. The message content is generic and is not based on any real-time choices made by a person. This dissertation describes the development and testing of an information system that uses context-aware algorithms to generate computer-tailored feedback messages to motivate healthy behavior. Such algorithms refer to an application's ability to adapt to changing circumstances and respond according to the context of use. The system developed for this dissertation can generate contextual feedback based on choices made by the user that have health consequences such as current location, pace, and activity, using inexpensive location, virtual and physical state sensors. This dissertation followed a Design Science paradigm of building an artifact, consisting of a smartphone and a server, and evaluating it through controlled field experiments. Subjects were recruited to wear accelerometer sensors, carry a GPS receiver and the smartphone. They followed instructions displayed on the phone as part of a course. At certain points they received dynamic contextualized feedback about choices they made while following instructions such as the use of stairs and elevator. The experiments helped understand the effects contextualized feedback have on short-term behavior and demonstrated the feasibility of using sensors to detect context.