Validation of nonmonotonic knowledge-based systems

Abstract

Nonmonotonic logical systems are logics in which the introduction of new axioms can invalidate old theorems. For practical applications, the authors restrict themselves to nonmonotonic systems that consist of facts and rules, and call them nonmonotonic knowledge-based systems (KBSs). Rules in a nonmonotonic KBS can be either default rules or regular rules. The syntax and semantics of these two types of rules are given. In contrast to other approaches, such as circumscription, default reasoning, assumption-based truth maintenance, and multi-valued logics, the authors directly extend the monotonic first order logic and its semantics. By using this method, one can define the validation problems for nonmonotonic KBSs similar to those for monotonic KBSs. Therefore, the validation tool already developed for monotonic KBSs can be extended to nonmonotonic KBSs.<>