Word Normalization Information Systems and Improved Learning Representation for Ontology Matching
Minxian Wang, Jing Peng · 2022 IEEE International Conference on Electrical Engineering, Big Data and Algorithms (EEBDA) · 2022
Biomedical ontology matching can find similar corresponds between biomedical ontologies to enable data integration. There are many ontology matching methods. Featurebased methods mainly focus on concept features to find mappings, ignoring semantic information and domain-specific morphology. On the other hand, learning representation methods think about the semantic information for hidden mappings but don't make full use of ontology structure. In this research, we propose a strategy for increasing performance by integrating the representation learning method with the feature-based method of additional word normalization. Our experiment is based on AgreementMakerLight (AML) and shows promising results.