header

GBKOM: A Generic framework for BK-based Ontology Matching

15 Pages Posted: 29 Sep 2020 Publication Status: Accepted

See all articles by Amina Annane

Amina Annane

University of Montpellier - Laboratory of Informatics, Robotics and Microelectronics (LIRMM)

Zohra Bellahsene

University of Montpellier - Laboratory of Informatics, Robotics and Microelectronics (LIRMM)

Abstract

BK-based matching exploits external background knowledge resources (BK) to ll the semantic gap between the ontologies to align. Existing BK-based matchers implement the indirect matching approach in their internal architecture, which makes any adaptation or reuse of the code dicult. Indeed, to improve a particular step in the BK-based matching process, it is necessary to code the whole process from scratch which requires a lot of time and eort. To overcome this issue, we propose a exible framework called Generic BK-based Matcher (GBKOM). GBKOM is an extension that can be added to any existing matcher. It is a congurable framework that implements the BK-based matching process, with a rich set of parameters making it customizable and suitable for performing experimental evaluations.

GBKOM has participated, with YAM++ as a direct matcher, in the OAEI 2017 and OAEI 2017.5 campaigns, where it has been successful on the biomedical benchmarks, and top ranked in several tasks. Furthermore, we have performed experiments with two other direct matchers (i.e., LogMap and LogMapLite) to show that the eectiveness of GBKOM is independent of the direct matcher used.

Keywords: Ontology matching, Ontology alignment, Background knowledge, Indirect matching, External resource, Anchoring, Derivation, Background knowledge selection

Suggested Citation

Annane, Amina and Bellahsene, Zohra, GBKOM: A Generic framework for BK-based Ontology Matching (September 22, 2020). Available at SSRN: https://ssrn.com/abstract=3697495 or http://dx.doi.org/10.2139/ssrn.3697495

Amina Annane (Contact Author)

University of Montpellier - Laboratory of Informatics, Robotics and Microelectronics (LIRMM) ( email )

163 rue Auguste Broussonnet
Montpellier
France

Zohra Bellahsene

University of Montpellier - Laboratory of Informatics, Robotics and Microelectronics (LIRMM) ( email )

163 rue Auguste Broussonnet
Montpellier
France

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
299
Downloads
29
PlumX Metrics