SEEK2: A Generalized Approach to Automatic Knowledge Base Refinement

Abstract

This paper describes an approach to knowledge base refinement, an important aspect of knowledge acquisition. Knowledge base refinement is characterized by the addition, deletion, and alteration of rule-components in an existing knowledge base, in an attempt to improve an expert system's performance. SEEK2 extends the capabilities of its predecessor rule refinement system. SEEK [1], In this paper we describe the progress we have made since developing the original SEEK program: (a) SEEK2 works with a more general class of knowledge bases than SEEK, (b) SEEK2 has an "automatic pilot " capability, i.e., it can, if desired, perform all of the basic tasks involved in knowledge base refinement without human interaction, (c) a metalanguage for knowledge base refinement has been specified which describes both domain-independent and domain-specific metaknowledge about the refinement process.