nerc.ac.uk

Ontology alignment based on word embedding and random forest classification

Nkisi-Orji, Ikechukwu ORCID: https://orcid.org/0000-0001-9734-9978; Wiratunga, Nirmalie ORCID: https://orcid.org/0000-0003-4040-2496; Massie, Stewart ORCID: https://orcid.org/0000-0002-5278-4009; Hui, Kit-jing ORCID: https://orcid.org/0000-0001-8383-7954; Heaven, Rachel ORCID: https://orcid.org/0000-0002-6172-4809. 2019 Ontology alignment based on word embedding and random forest classification. In: Berlingerio, Michele; Bonchi, Francesco; Gärtner, Thomas, (eds.) Machine learning and knowledge discovery in databases. Springer, 557-572. (Lecture notes in computer science, 11051).

Before downloading, please read NORA policies.
[thumbnail of Ontology_alignment_based_on_word_embedding_and_random_forest_classification.pdf]
Preview
Text
Ontology_alignment_based_on_word_embedding_and_random_forest_classification.pdf - Accepted Version

Download (528kB) | Preview

Abstract/Summary

Ontology alignment is crucial for integrating heterogeneous data sources and forms an important component for realising the goals of the semantic web. Accordingly, several ontology alignment techniques have been proposed and used for discovering correspondences between the concepts (or entities) of different ontologies. However, these techniques mostly depend on string-based similarities which are unable to handle the vocabulary mismatch problem. Also, determining which similarity measures to use and how to effectively combine them in alignment systems are challenges that have persisted in this area. In this work, we introduce a random forest classifier approach for ontology alignment which relies on word embedding to discover semantic similarities between concepts. Specifically, we combine string-based and semantic similarity measures to form feature vectors that are used by the classifier model to determine when concepts match. By harnessing background knowledge and relying on minimal information from the ontologies, our approach can deal with knowledge-light ontological resources. It also eliminates the need for learning the aggregation weights of multiple similarity measures. Our experiments using Ontology Alignment Evaluation Initiative (OAEI) dataset and real-world ontologies highlight the utility of our approach and show that it can outperform state-of-the-art alignment systems.

Item Type: Publication - Book Section
Digital Object Identifier (DOI): 10.1007/978-3-030-10925-7_34
ISBN: 9783030109240
Additional Keywords: Ontology alignment, Word embedding, Machine classification, Semantic web
NORA Subject Terms: Computer Science
Date made live: 31 Mar 2023 08:57 +0 (UTC)
URI: https://nora.nerc.ac.uk/id/eprint/534286

Actions (login required)

View Item View Item

Document Downloads

Downloads for past 30 days

Downloads per month over past year

More statistics for this item...