Micro-context recognition of sedentary behaviour using smartphone

Embedded sensors of smartphone provides a unique opportunity to recognize the micro-context of sedentary behaviour. In this paper, we present our research findings on how to recognize micro-contexts by utilizing on board sensors of smartphone. Our proposed approach consists of two stages process. First, we recognize the situation of a person to be either stationery or moving. If stationary, then high probability to be sedentary, in which we can then find micro details about the current context. Second, we process environmental sound and recognize the person's micro-context such as watching television, working on computers or relaxing. Furthermore, we also provide the lifestyle analytics over cloud computing infrastructure to make it available anywhere and anytime for self-management purpose. We developed an initial working prototype to evaluate the applicability of our approach in a real-world scenario.