The role of machine learning in behavior change
There is no one-size-fits-all solution to health behavior change. Some people respond to gentle nudges, others benefit more from structured programs, some take pride in being a Do-It-Yourselfer, others make more progress with a coach. Some can only change behavior if it is presented in the form of a game, others respond to social interactions.
At Fitbit R&D, we are interested in answering some fundamental questions: what motivates a person to lead a healthier life, what is a real or perceived barrier for change, and how Fitbit can help increase users' ability to make those changes. Answering these questions requires a multi-disciplinary approach. We take advantage of health science, behavior science, behavioral economics, and machine learning to deliver interventions. In this talk, we will review how machine learning is a key capability in making any solution smart and more personalized.