Foundation and application of knowledge base verification

Abstract

Anomalies such as redundant, contradictory, and deficient knowledge in a knowledge base are symptoms of probable errors. Detecting anomalies is a well-established method for verifying knowledge-based systems. Although many tools have been developed to perform anomaly detection, several important issues have been neglected, especially the theoretical foundations and computational limitations of anomaly detection methods, and analyses of the utility of such tools in practical use. This article addresses these issues by presenting a theoretical foundation of anomaly detection methods, and by presenting empirical results obtained in applying one anomaly detection tool to perform verification on five real-world knowledge-based systems. the techniques presented apply specifically to verifying rule-based knowledge bases without numerical certainty measures. © 1994 John Wiley & Sons, Inc.