Distance semantics for database repair

  • Authors:
  • Ofer Arieli;Marc Denecker;Maurice Bruynooghe

  • Affiliations:
  • Department of Computer Science, The Academic College of Tel-Aviv, Tel-Aviv, Israel;Department of Computer Science, Katholieke Universiteit Leuven, Leuven, Belgium;Department of Computer Science, Katholieke Universiteit Leuven, Leuven, Belgium

  • Venue:
  • Annals of Mathematics and Artificial Intelligence
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

In many scenarios, a database instance violates a given set of integrity constraints. In such cases, it is often required to repair the database, that is, to restore its consistency. A primary motif behind the repairing approaches is the principle of minimal change, which is the aspiration to keep the recovered data as faithful as possible to the original (inconsistent) database. In this paper, we represent this qualitative principle quantitatively, in terms of distance functions and some underlying metrics, and so introduce a general framework for repairing inconsistent databases by distance-based considerations. The uniform way of representing repairs and their semantics clarifies the essence behind several approaches to consistency restoration in database systems, helps to compare the underlying formalisms, and relates them to existing methods of defining belief revision operators, merging data sets, and integrating information systems.