Chapter

Ontology Matching

pp 149-197

Matching Strategies

  • Jérôme EuzenatAffiliated withINRIA and LIG
  • , Pavel ShvaikoAffiliated withInformatica Trentina SpA, while at Department of Engineering and Computer Science (DISI), University of Trento, while at Web of Data, Bruno Kessler Foundation - IRST

* Final gross prices may vary according to local VAT.

Get Access

Abstract

The basic techniques presented in Chap. 5 and the global techniques provided in Chap. 6 are the building blocks on which a matching system is built. Once the similarity or dissimilarity between ontology entities is available, the alignment remains to be computed. This involves more comprehensive treatments. In particular, the following aspects of building a working matching system are considered in this chapter:
  • preparing, if necessary, to handle large scale ontologies (Sect. 7.1.1),

  • organising the combination of various similarities or matching algorithms (Sect. 7.2),

  • exploiting background knowledge sources (Sect. 7.3),

  • aggregating the results of the basic methods in order to compute the compound similarity between entities (Sect. 7.4),

  • learning matchers from data (Sect. 7.5) and tuning them (Sect. 7.6),

  • extracting alignments from the resulting (dis)similarity: indeed, different alignments with different characteristics may be extracted from the same (dis)similarity (Sect. 7.7),

  • improving alignments through disambiguation, debugging and repair (Sect. 7.8).