An Architecture for Context Prediction

Context-based interaction is one of the building blocks of Pervasive Computing. During the last five years, a number of seminal publications on the recognition of current context, that is the current situation within actions are taking place, have been written within this field.
This work tackles the next logical step after the recognition of the current context: the prediction of future contexts. Proactivity, which, in contrast to reactivity, means the execution of actions for reaching a defined goal before external events trigger them, is one of the most distinguishing human abilities. It is an obvious reason why human assistants can not yet be replaced by their digital counterparts in numerous areas. By applying proactivity in computer systems, they can adapt to expected future events in advance and therefore provide better services to users. This book represents the first systematic work on the integration of context-based interaction with proactivity.
The general concept is the prediction of abstract contexts - in contrast to the autonomous prediction of individual aspects like the geographical position of the user. This work analyzes prerequisites for user-centered prediction of context and presents an architecture for autonomous, background context recognition and prediction, building upon established methods for data based prediction. Particular attention is turned to implicit user interaction to prevent disruptions of users during their normal tasks and to continuous adaption of the developed systems to changing conditions. Another considered aspect is the economical use of resources to allow the integration of context prediction into embedded systems. The developed architecture has been implemented as a flexible software framework and evaluated with recorded real-world data from everyday situations.