Research Article
Matching Ontologies Through Siamese Neural Network
@INPROCEEDINGS{10.1007/978-3-030-89814-4_52, author={Xingsi Xue and Chao Jiang and Hai Zhu}, title={Matching Ontologies Through Siamese Neural Network}, proceedings={Mobile Multimedia Communications. 14th EAI International Conference, Mobimedia 2021, Virtual Event, July 23-25, 2021, Proceedings}, proceedings_a={MOBIMEDIA}, year={2021}, month={11}, keywords={Ontology matching Siamese neural networks OAEI}, doi={10.1007/978-3-030-89814-4_52} }- Xingsi Xue
Chao Jiang
Hai Zhu
Year: 2021
Matching Ontologies Through Siamese Neural Network
MOBIMEDIA
Springer
DOI: 10.1007/978-3-030-89814-4_52
Abstract
Ontology, the kernel technique of Semantic Web (SW), formally names the domain concepts and their relationships. However, as the ontologies are created and developed by different domain experts and communities, a concept may be named in various ways, bringing about the concept heterogeneity problem. To solve the problem, in this paper, a Siamese Neural Network (SNN)-based Ontology Matching Technique (OMT) is proposed, which is able to improve the matching efficiency by using a part of Reference Alignment (RA) to decrease the training time and improve the quality of matching results by using a logic reasoning approach to remove the conflict correspondences. The experimental results demonstrate that SNN-based OMT can determine high-quality alignment which outperforms the state-of-the-art OMTs.
