Publication Type: Conference Paper
Source: Proceedings Ninth International Conference. Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU 2002, July 1-5, 2002, Annecy, France), Volume III, p.1655-1662 (2002)
The main goal of this paper is to present an approach to restore consistency in systems of fuzzy gradual rules. The approach involves the use of similarity relations to transform potentially inconsistent rules by making them more imprecise. Namely, we propose to replace their consequents by their image through a suitable similarity relation, in such a way that it results into more imprecise rules but avoiding consistency problems.