Correcting queries for XML

  • Authors:
  • Sara Cohen;Tali Brodianskiy

  • Affiliations:
  • The Selim and Rachel Benin School of Computer Science and Engineering, The Hebrew University of Jerusalem, Jerusalem 91904, Israel;Faculty of Industrial Engineering and Management, Technion-Israel Institute of Technology, Technion City, Haifa, Israel

  • Venue:
  • Information Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

It has been observed that queries over XML data sources are often unsatisfiable. Unsatisfiability may stem from several different sources, e.g., the user may be insufficiently familiar with the labels appearing the documents, or may not be intimately aware of the hierarchical structure of the documents. To deal with query and document mismatches, previous research has considered returning answers that maximally satisfy (in some sense) the query, instead of only returning strictly satisfying answers. However, this breaks the golden database rule that only strictly satisfying answers are returned when querying. Indeed, the relationship between the query and answers is no longer clear, when unsatisfying answers are returned. To reinstate the golden database rule, this article proposes a framework for automatically correcting queries over XML. This framework generates similar satisfiable queries, when the user query is unsatisfiable. The user can then choose a satisfiable query of interest, and receive exactly satisfying answers to this query.