Current mobile phone and wearable sensor technologies allow us to track a person's physical activities throughout their daily lives. In many circumstances, however, there are strong incentives for users to trick the activity recognition systems into detecting activities that are different from the actual ones. The current activity recognition technology is only reliable in normal conditions, and thus vulnerable to such behavior. The aim of this project is to develop a methodology that enables smartphone-based activity recognition to overcome this limitation.
Supported by P20MH09031 (Principal Investigator: David C. Mohr, Ph.D.)