Integrating qualitative reasoning with large-scale knowledge bases provides new challenges. This paper outlines work in progress on developing a new model formulation system to support qualitative reasoning via compositional modeling that can operate in an environment with over a million available facts. Three ideas are discussed: Exploiting microtheories for modeling, using non-monotonic inference, and speculative inference. An implemented algorithm which runs on QP benchmark examples is outlined, and planned extensions are discussed.
|Title of host publication
|Proceedings of QR10
|Published - 2010
|24th International Workshop on Qualitative Reasoning - Portland, OR
Duration: Aug 1 2010 → …
|24th International Workshop on Qualitative Reasoning
|8/1/10 → …