Reference Hub2
Rough Set Based Ontology Matching

Rough Set Based Ontology Matching

Saruladha Krishnamurthy, Arthi Janardanan, B Akoramurthy
Copyright: © 2018 |Volume: 5 |Issue: 2 |Pages: 23
ISSN: 2334-4598|EISSN: 2334-4601|EISBN13: 9781522547020|DOI: 10.4018/IJRSDA.2018040103
Cite Article Cite Article

MLA

Krishnamurthy, Saruladha, et al. "Rough Set Based Ontology Matching." IJRSDA vol.5, no.2 2018: pp.46-68. http://doi.org/10.4018/IJRSDA.2018040103

APA

Krishnamurthy, S., Janardanan, A., & Akoramurthy, B. (2018). Rough Set Based Ontology Matching. International Journal of Rough Sets and Data Analysis (IJRSDA), 5(2), 46-68. http://doi.org/10.4018/IJRSDA.2018040103

Chicago

Krishnamurthy, Saruladha, Arthi Janardanan, and B Akoramurthy. "Rough Set Based Ontology Matching," International Journal of Rough Sets and Data Analysis (IJRSDA) 5, no.2: 46-68. http://doi.org/10.4018/IJRSDA.2018040103

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Ontologies enriches the knowledge and add meaning to the data residing in semantic web. Ontology matching identifies concepts for matching in source ontologies to the target ontologies to eliminate heterogeneities. Despite using similarity measures for identifying similar concepts, the ontology matching systems fails to handle uncertainty. This paper proposes a rough set based ontology matching system to handle uncertainty which aims (i)to optimize concepts considered for matching by using concept type classification(ii) to use rough set concepts using indiscernibility relations and reducts (iii) to apply criterion of realism – a decision making under uncertainty criteria. The experiments conducted in the OAEI benchmark data sets, RSOM system yielded an increase of 8% in precision. The combined approach of using reduct and indiscenibilty relations reduces the number of concepts considered for matching among uncertain entities to about 70% in comparison to the existing systems and increases the accuracy of results by using criterion of realism.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.