A KDD framework to support database audit

  • Authors:
  • Jean-françois Boulicaut

  • Affiliations:
  • Institut National des Sciences Appliquées de Lyon, LISI Bâtiment 501, F-69621 Villeurbanne Cedex, France E-mail: jean-francois.boulicaut@insa-lyon.fr

  • Venue:
  • Information Technology and Management
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

Understanding data semantics from real-life databases is considered following an audit perspective: it must help experts to analyse what properties actually hold in the data and support the comparison with desired properties. This is a typical problem of knowledge discovery in databases (KDD) and it is specified within the framework of Mannila and Toivonen where data mining consists in querying theories e.g., the theories of approximate inclusion dependencies. This formalization enables us to identify an important subtask to support database audit as well as a generic algorithm. Next, we consider the DREAM relational database reverse engineering method and DREAM heuristics are revisited within this new setting.