Abstract
Capturing contextual information, especially higher-level contexts enables systems to understand and predict the behaviour of a mobile user. This kind of information is mostly implicit contexts, which abstract a complex state of a situation and can only partly be captured by employing sensors. Higher-level context has the potential to make user-application interaction richer, simpler, and more intuitive. However, composing higher-level contexts from explicit, atomic contexts requires complex and painstaking reasoning procedures to resolve uncertainty due to inconsistent sensor readings and incomplete information. In this paper, we introduce a new approach that assists application developers to take higher-level contexts into account without the need to know the details of atomic contexts. To demonstrate our approach, we will introduce the Context-Aware E-Pad (CAEP) we have designed and implemented. CAEP “observes” the user making a decision and associates with it a set of atomic contexts. Likewise, correlated decisions are mapped to similar associations. By repeatedly “observing” the user making decisions, CAEP can predict user’s wishes and behaviours.
Users
Please
log in to take part in the discussion (add own reviews or comments).