Dargie, Waltenegus (Chair of Computer Networks, Faculty of Computer Science, Technical University of Dresden, D-01062 Dresden, Germany)
An essential aspect of Mark Weiser's vision for ubiquitous computing is that computers become "invisible", demanding very little attention of their users. For this to happen, computers should be able to establish a shared understanding of the conceptual and social settings in which they operate, i.e., they should be context aware. This paper addresses one aspect of contextawareness, namely the management of context rules in mobile and wearable devices. Context rules are employed to predict higher-level human situations (activities), and to actuate desirable operations which are suitable to these situations. Because context rules are application specific, they are defined by application developers and users. So far, little consideration is given in the literature to take the habit (experience) of users into account to dynamically generate and manage context rules. In this paper we will provide a framework to associate decision events (signifying the actions performed by a user) with numerous context types which describe a situation of interest. From an aggregate of decision-context associations we generate context rules which enable mobile devices to proactively provide useful services.