Pervasive Computing is a new research area at the intersection between
human/computer interaction, embedded and distributed systems and 
networking technology. Its aim - the disappearance of computer technology 
into the periphery of daily life - necessitates an adaption of systems to 
the respective context in which they are used. Context-based interaction 
is therefore one of the building blocks of Pervasive Computing. Within the
last five years, a number of seminal publications on the recognition of
current context from a combination of different sensors have been written.

The next logical step is the prediction of future contexts, with the 
general concept of predicting abstract contexts to allow computer systems 
to proactively prepare for future situations. This high-level context
prediction allows - in contrast to the autonomous prediction of individual
aspects like the geographical position - to consider patterns and
interrelations in the user behavior which are not apparent at the lower 
levels of raw sensor data. In this talk, user-centered prediction of 
context is analyzed and an architecture for autonomous, background context
recognition and prediction is presented, building upon established methods 
for data based prediction. Especial attention is turned to implicit user
interaction to prevent disruptions of users during their normal tasks, to
continuous adaption of systems to changed conditions, and to economical 
use of resources.