Language-model-based ranking for queries on RDF-graphs

  • Authors:

  • Shady Elbassuoni;Maya Ramanath;Ralf Schenkel;Marcin Sydow;Gerhard Weikum

  • Affiliations:

  • Max-Planck Institute for Informatics, Saarbrucken, Germany;Max-Planck Institute for Informatics, Saarbrucken, Germany;Max-Planck Institute for Informatics, Saarbrucken, Germany;Polish-Japanese Institute of Information Technology, Warsaw, Poland;Max-Planck Institute for Informatics, Saarbrucken, Germany

  • Venue:
  • Proceedings of the 18th ACM conference on Information and knowledge management
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The success of knowledge-sharing communities like Wikipedia and the advances in automatic information extraction from textual and Web sources have made it possible to build large "knowledge repositories" such as DBpedia, Freebase, and YAGO. These collections can be viewed as graphs of entities and relationships (ER graphs) and can be represented as a set of subject-property-object (SPO) triples in the Semantic-Web data model RDF. Queries can be expressed in the W3C-endorsed SPARQL language or by similarly designed graph-pattern search. However, exact-match query semantics often fall short of satisfying the users' needs by returning too many or too few results. Therefore, IR-style ranking models are crucially needed. In this paper, we propose a language-model-based approach to ranking the results of exact, relaxed and keyword-augmented graph pattern queries over RDF graphs such as ER graphs. Our method estimates a query model and a set of result-graph models and ranks results based on their Kullback-Leibler divergence with respect to the query model. We demonstrate the effectiveness of our ranking model by a comprehensive user study.