Intervention at Appropriate Timing to Increase Steps Based on FBM

Abstract

To increase the steps users take, the paper proposes a method founding on the Fogg Behavior Model (FBM). The method estimates the best timing to intervene with them. Exercise is essential for maintaining people's health. However, the amount of time spent in sedentary activities has been increasing in recent years. People's health is at risk. The FBM explains that people are inclined to take action when they have high motivation and ability. It also points out messages to provide triggers to increase steps vary with motivation and ability. The proposed method obtains step count data for each day of the week through an activity tracker. The data are processed to identify times when the user's step count is steadily increasing. They are regarded as high-ability periods. User motivation is obtained from a machine learning model.  The model is constructed from the physical activity data through the user's activity tracker collected in the past and questionnaires on the user's motivation for exercise. The method provides intervention according to the user's motivation during the high-ability periods. The proposed method is examined in an experiment. The results have shown the proposed method improves the steps significantly.