Database design with equality-generating dependencies

  • Authors:
  • Junhu Wang

  • Affiliations:
  • School of Information Technology, Griffith University, Gold Coast, Australia

  • Venue:
  • DASFAA'05 Proceedings of the 10th international conference on Database Systems for Advanced Applications
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

In relational database systems, traditional normalization techniques (eg, BCNF, 4NF) remove data redundancies from a single relation, but can not detect and remove redundancies across multiple relations. However, redundancies among multiple relations are abundant especially in integrated databases. In this paper, we first propose to detect such data redundancies using equality-generating dependencies (EGDs) and propose an extended normal form (ENF) of relational database schema with respect to EGDs. We show that a database has no potential data redundancies with respect to EGDs if and only if the schema is in ENF. For a common special class of EGDs, we provide a set of sound and complete inference rules. A normalization process is presented to losslessly transform a relational database schema to one in ENF. We then extend our EGDs and ENF to XML data, and show how similar data redundancy problems can be detected for data-centric XML documents.