Modeling uncertainty in context-aware computing

Uncertainty always exists as an unavoidable factor in any pervasive context-aware applications. This is mostly caused by the imperfectness and incompleteness of data. In this paper, we propose a novel approach to model the uncertain context. Our context model is a combination of two modeling methods: probabilistic models for capturing the uncertain information and ontology for facilitating knowledge reuse and sharing. Such combination of probabilistic models and ontology facilitates the sharing and reuse over similar domains of not only the logical knowledge but also the uncertain knowledge. Besides, we also support the uncertain reasoning in context-aware applications in a flexible and adaptive manner.