Abductive Reasoning

Abductive reasoning has widespread application in natural language processing systems. For example, Guessoum, Black, and Gallagher [6] describe the use of abduction for pronoun resolution. Most of the existing NLP systems use linguistic constraints for eliminating candidate referents, but it is widely recognized that non-linguistic knowledge is required to resolve ambiguities in general (c.f. the textbook example `If the baby doesn't thrive on cows' milk, boil it'). More interesting for our testbed is the work of Quaresma and Lopes [13] on abduction of plans and intentions in dialogues.

Hustadt [7] proposes an abductive reasoning method for disjunctive logic programs that has been integrated into MOTEL.


next up previous
Next: Reason Maintenance Up: The MOTEL Knowledge Representation System Previous: Non-monotonic Reasoning [an error occurred while processing this directive]