Candidate reduction and alignment improvement techniques used in aligning ontologies

  • Authors:

  • Watson Wei Khong Chua;Angela Eck Soong Goh

  • Affiliations:

  • Nanyang Technological University, Singapore;Nanyang Technological University, Singapore

  • Venue:
  • Proceedings of the 11th International Conference on Information Integration and Web-based Applications & Services
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Ontology alignment is required to enable interoperability between applications using different ontologies. The alignments generated allow knowledge to be shared between semantically equivalent concepts in two different ontologies. Some applications carry out the ontology alignment process offline and the accuracy of the alignments generated is much more important than the performance of the process. In other applications that perform the ontology alignment process online, the performance is as important as the accuracy of the alignments generated because the process needs to be fast. In this paper, we present techniques used to reduce the number of candidates for matching and techniques used to improve the accuracy of alignments in the ontology alignment process. Candidate reduction methods compare entities in the source ontology to a subset of entities in the target ontology by filtering entities which are unlikely to be matched instead of extensively comparing all pairs of entities belonging to two different ontologies. These techniques reduce the time taken to align the ontologies and are categorised depending on whether they perform candidate reduction using structural methods, external resources or iterative methods. On the other hand, techniques to improve the accuracy of alignments output a set of alignments with greater accuracy given an initial set of alignments. Some of these techniques are constraint-based while others are not.