Prolog Meta-Interpreters for Rule-Based Inference Under Uncertainty

Uncertain facts and inexact rules can be represented andprocessed in standard Prolog through meta-interpretation. Thisrequires the specification of appropriate parsers and beliefcalculi. We present a meta-interpreter that takes a rule-basedbelief calculus as an external variable. The certainty-factorscalculus and a heuristic Bayesian belief-update model are thenimplemented as stand-alone Prolog predicates. These, in turn,are bound to the meta-interpreter environment through second-orderprogramming. The resulting system is a powerfulexperimental tool which enables inquiry into the impact ofvarious designs of belief calculi on the external validity ofexpert systems. The paper also demonstrates the (well-known)role of Prolog meta-interpreters in building expert systemshells