PSemRef: personalized query reformulation based on user preferences

  • Authors:
  • Thiago Arruda;Damires Souza;Ana Carolina Salgado

  • Affiliations:
  • Federal University of Pernambuco, PE, Brazil;Federal Institute of Education, PB, Brazil;Federal University of Pernambuco, PE, Brazil

  • Venue:
  • Proceedings of the 12th International Conference on Information Integration and Web-based Applications & Services
  • Year:
  • 2010

Quantified Score

Hi-index 0.01

Visualization

Abstract

One key issue for query answering in dynamic distributed environments is the reformulation of a query posed at a peer into another one over a target peer. Making use of the semantics underlying a set of correspondences between peer schemas' elements, the SemRef approach has been developed as a means to enhance such process. Nevertheless, while such approach is able to provide users with a set of expanded answers, it lacks to tailor query results according to user's preferences on the different existing semantic correspondences options. In this paper, we address the issue of personalizing query results, in such a way that users may choose which types of semantic correspondences are important to their queries as well as the priority order in which these correspondences should be applied. More specifically, we address query personalization at reformulation time, producing a ranked set of answers according to user's preferences. We present the principles underlying our approach, examples illustrating how they work and some experimental results.