Adding similarity-based reasoning capabilities to a Horn fragment of possibilistic logic with fuzzy constants

PLFC is a 0rst-order possibilistic logic dealing with fuzzy constants and fuzzily restricted quanti0ers. The refutation proof method in PLFC is mainly based on a generalized resolution rule which allows an implicit graded uni0cation among fuzzy constants. However, uni0cation for precise object constants is classical. In order to use PLFC for similarity-based reasoning, in this paper we extend a Horn-rule sublogic of PLFC with similarity-based uni0cation of object constants. The Horn-rule sublogic of PLFC we consider deals only with disjunctive fuzzy constants and it is equipped with a simple and e5cient version of PLFC proof method. At the semantic level, it is extended by equipping each sort with a fuzzy similarity relation, and at the syntactic level, by fuzzily “enlarging” each non-fuzzy object constant in the antecedent of a Horn-rule by means of a fuzzy similarity relation. c © 2003 Elsevier B.V. All rights reserved.