A preference query model based on a fusion of local orders

  • Authors:
  • Patrick Bosc;Olivier Pivert;Grégory Smits

  • Affiliations:
  • Irisa, Enssat, IUT, University of Rennes 1, Lannion Cedex, France;Irisa, Enssat, IUT, University of Rennes 1, Lannion Cedex, France;Irisa, Enssat, IUT, University of Rennes 1, Lannion Cedex, France

  • Venue:
  • ECSQARU'11 Proceedings of the 11th European conference on Symbolic and quantitative approaches to reasoning with uncertainty
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we define an approach to database preference queries based on the fusion of local orders. The situation considered is that of queries involving incommensurable partial preferences, possibly associated with scoring functions. The basic principle is to rank the tuples according to each partial preference, then to merge the local orders obtained, using a linear function for aggregating the local scores attached to the tuples. Basically, a local score expresses the extent to which a tuple is strictly better than many others and not strictly worse than many others with respect to the partial preference attached to a given attribute. This model refines Pareto order for queries of the Skyline type.