Critiquing with Multiple Criteria: Conflict Detection and Resolution

Dan Rochowiak, John Rogers, and Sherri Messime

Building a taxonomy of conflicts is difficult. At smaller grain sizes, the types of conflicts are as diverse as the specific domains in which they occur. At larger grain sizes, the types of conflict are as diverse as the theories of conflict detection and management. This diversity should not be surprising. Conflicts .reflect the differing approaches that individuals have to common problems, and the approaches of the individuals are shaped by their specific areas of expertise. Our approach accepts this sort of diversity, but attempts to structure the process of conflict recognition and management through the notions of parameter values and criteria that do not always demand complete satisfaction in a rigidly defined system of criteria. Our approach focuses on the "fuzziness" of reasoning about planning and decision making, and the construction of critiques based on degrees of satisfaction. It should be noted the account of "fuzziness" that we use is not the same as that given in standard fuzzy logic. Rather our focus is on the degree to which less than precise and exact parameter values can be appraised by less than precise and exact criteria. The system is currently being developed in the field of composite materials design and manufacturing.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.