Abstract
This paper presents a writing state classification technique by a sedentary behavior sensing with a sensor equipped chair, which wirelessly monitors user's states such as his/her behaviors or physiological and psychological conditions, for the estimation of the user's subjective difficulty of a studying problem on a desktop. Avoiding user's discomfort, four load cells are equipped behind the seat plate of a regular office chair. The system simply measures the weight and the CoP (Center of Pressure) of the seat. These data are divided into time segments, which are labeled into four sedentary body sway primitives by the decision tree algorithm, and then the behaviors are derived from the number and the pitch of those primitives in a certain interval. System evaluation experiments resulted that it achieved more than 80% of labeling accuracy for the untrained users. We finally conducted a user experiment that shows its potential for desk work applications. The result of the experiment showed that our proposed system could classify user's states into writing state and the other states such as reading a document and watching movie at the classification rate of 87%.
