A method for learning part-whole relations

  • Authors:
  • Willem Robert van Hage;Hap Kolb;Guus Schreiber

  • Affiliations:
  • TNO Science & Industry Delft;TNO Science & Industry Delft;Vrije Universiteit Amsterdam

  • Venue:
  • ISWC'06 Proceedings of the 5th international conference on The Semantic Web
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Part-whole relations are important in many domains, but typically receive less attention than subsumption relation. In this paper we describe a method for finding part-whole relations. The method consists of two steps: (i) finding phrase patterns for both explicit and implicit part-whole relations, and (ii) applying these patterns to find part-whole relation instances. We show results of applying this method to a domain of finding sources of carcinogens.