Context-aware systems gained huge popularity in recent
years due to rapid evolution of personal mobile devices. Equipped with
variety of sensors, such devices are sources of a lot of valuable information
that allows the system to act in an intelligent way. However, the
certainty and presence of this information may depend on many factors
like measurement accuracy or sensor availability. Such a dynamic
nature of information may cause the system not to work properly or
not to work at all. To allow for robustness of the context-aware system
an uncertainty handling mechanism should be provided with it. Several
approaches were developed to solve uncertainty in context knowledge
bases, including probabilistic reasoning, fuzzy logic, or certainty
factors. In this paper, we present a representation method that combines
strengths of rules based on the attributive logic and Bayesian networks.
Such a combination allows efficiently encode conditional probability distribution
of random variables into a reasoning structure called XTT2.
This provides a method for building hybrid context-aware systems that
allows for robust inference in uncertain knowledge bases.