Classification of daily activities based on the amount of social and physical activity for behavioral change toward wellbeing

In this study, we classify daily activities by focusing on the relationship between social and physical activities. By self-reflection through lifelogging, we aim to discover ways of spending and indicators that lead to physical and mental health, such as increased satisfaction and reduced fatigue. We classify daily activities based on the amount of physical activity obtained from a smartwatch and the amount of social activity obtained from a neck-worn face-counting device, and obtain insights into the relationship between social and physical activities. First, we conduct a hierarchical clustering of daily activities (8 types, 37 scenes, 10 minutes each) of an individual, which are then are classified into four groups. Then, the amount of physical activity and social activity are plotted on a two-dimensional plane to analyze the tendency. In addition, we plot the activities of a group of three participants on the same space and analyzed the tendency of each activity. Although there are differences in the amount of physical activity and social activity among the individuals, they tend to fall into the groups. Moreover, we visualize the transition of the amount of physical activity and social activity of one participant throughout a half-day. We are interested in the time-series rhythm and balance of the person’s activities. We believe that our visualization will suggest activities for individuals based on their state of physical and social (mental) health.