Hybridising events and knowledge in an infrastructure for context-adaptive systems

Event-based systems are a popular substrate for distributing information derived from sensors to be used in driving adaptive behaviour. We argue that event systems only provide a poor model of context, and that a hybrid approach that uses events to populate and maintain a knowledge base provides a more stable solution. The inherent uncertainties imply that traditional knowledgebased system techniques are extended to deal with more uncertain reasoning. We discuss our plans for additional work in analysing and programming autonomic behaviours with this architecture.