Optimizing Tableau Reasoning in ALC Extended with Uncertainty

There has been an increased interest in recent years to incorporate uncertainty in Description Logics (DLs), and a number of proposals have been put forward for modeling uncertainty in DL frameworks. While much progress has been made on syntax, semantics, and query processing issues, optimizing queries in this context has received little attention. In this paper, we study query processing for a tableau-based DL framework with uncertainty and focus on optimization of resolution of certainty inequality constraints, obtained from a translation in query processing phase. We develop a running prototype which evaluates DL knowledge bases with ABoxes and TBoxes annotated with uncertainty parameters and computes the corresponding semantics encoded as a set of constraints in the form of linear and/or nonlinear inequations. We also explore various existing and new opportunities for optimizing the reasoning procedure in this context. Our experimental evaluation indicates that the optimization techniques we considered result in improved efficiency significantly.