Can You Form Healthy Habit? Predicting Habit Forming States through Mobile Phone

Health-compromising behaviors are difficult to change since people do not behave in accordance with their intention. This paper aims at studying the extent to which a person’s healthy habit forming process can be affected by mobile phone usage. We propose a novel healthy habit forming states predicting framework using mobile phone platform. First we present a definition for the healthy habit forming process consisting of several states. We define the social intervention types and user context data which are extract-ed from mobile phone sensor data. Then we make use of machine learning methods to study the correlation between these data and healthy habit forming states. Specifically, a predicting model called Habits Factor Graph(HaFG) is pro-posed to predict the habit forming states. To evaluate our work, an Android based prototype system is implemented. Experimental results show that the healthy habit forming states are predicted possibly from user context information with a fairly good accuracy (around 67%)