Towards generating causal explanation: qualitative simulation withassociation mechanism to quantitative information

摘要

One of the central issues of qualitative reasoning is generatingcausal explanation in response to a user's query and the modelconstruction. Constructing an adequate model is crucial in generatingexplanation in case of large-scale physical systems. This paperdescribes new mechanisms in the model construction environment, usingthe causal ordering algorithm and qualitative conversion of numericaldata. Our approach involves two key ideas: a qualitative model isconstructed from a unique quantitative model, and simulated qualitativebehaviors are compared with numerical data. The result of applying themechanisms to power plants is discussed