Improving Sleep-Wake Schedule Using Sleep Behavior Visualization and a Bedtime Alarm
We demonstrate usage of smartphones and smartwatches to automatically recognize a person’s behavioral context in-the-wild. In our setup, subjects use their own personal phones and engage in regular behavior in their natural environments. Our system fuses complementary information from multi-modal sensors and simultaneously recognizes many contextual attributes from diverse behavioral domains, including work and leisure activities, body movement, transportation, and more. Our large scale labeled dataset is publicly available.