Automated Re nement of First - Order Horn - Clause Domain Theories

Knowledge acquisition is a di cult, error-prone, and time-consuming task. The task of automatically improving an existing knowledge base using learningmethods is addressed by the class of systems performing theory re nement. This paper presents a system, Forte (First-Order Revision of Theories from Examples), which re nes rst-order Horn-clause theories by integrating a variety of di erent revision techniques into a coherent whole. Forte uses these techniques within a hill-climbing framework, guided by a global heuristic. It identi es possible errors in the theory and calls on a library of operators to develop possible revisions. The best revision is implemented, and the process repeats until no further revisions are possible. Operators are drawn from a variety of sources, including propositional theory re nement, rst-order induction, and inverse resolution. Forte is demonstrated in several domains, including logic programming and qualitative modelling.