On integrity constraints

  • Authors:

  • Raymond Reiter

  • Affiliations:

  • University of Toronto, Toronto, Ontario, Canada

  • Venue:
  • TARK '88 Proceedings of the 2nd conference on Theoretical aspects of reasoning about knowledge
  • Year:
  • 1988

Quantified Score

Hi-index 0.00

Visualization

Abstract

We address the concept of a static integrity constraint as it arises in databases and Artificial Intelligence knowledge representation languages. Such constraints are meant to characterize the acceptable states of a knowledge base, and are used to enforce these legal states. We adopt the perspective that a knowledge base is a set of first order sentences, but argue, contrary to the prevailing view, that integrity constraints are epistemic in nature. Rather than being statements about the world, constraints are statements about what the knowledge base can be said to know. We formalize this notion in the language KFOPCE due to Levesque and define the concept of a knowledge base satisfying its integrity constraints. We investigate constraint satisfaction for closed world knowledge bases. We also show that Levesque's axiomatization of KFOPCE provides the correct logic for reasoning about integrity constraints. Finally, we show how to determine whether a knowledge base satisfies its constraints for a restricted, but important class of knowledge bases and constraints.