Classification may not be reliable for several reasons: noise in the data, insufficient input information, overlapping distributions and sharp definition of classes. Faced with K possibilities a decision support system may still be useful in such cases if instead of classification elimination of improbable classes is done. Eliminators may be constructed using classifiers assigning new cases to ...

In the past, little attention has been paid to critiquing in domains in which reasoning, planning and action are interleaved, or in which actions can change the agent's knowledge in ways that aaect subsequent reasoning and planning. This paper discusses the demands these characteristics make on a critiquing system. The goal-directed approach to critiquing that I am developing interacts with a p...