Using activity theory to identify relevant context parameters

Abstract

One of the most important aspects of ubiquitous computing is context-awareness. In this paper, we adopt an interactional perspective on context: 1) Something is context because it is used for adapting the interaction between the human and the current system. 2) Activity is central to context. 3) Context differs in each occasion of the activity. Based on this understanding, this paper proposes an Activity Theory based method which attempts to answer the following questions: how to analyze activity for context-awareness, and how to identify relevant context parameters. This method includes two steps: 1) Decomposing activity into actions, which we take as units for identifying context parameters, by using Activity Theory’s hierachical structure of activity. 2) Identifying relevant context parameters for each action by our extended Activity Theory’s framework. Finally, this paper gives an outlook how this method can be used in designing context-aware pedestrian wayfinding services.