Biomedical ontology alignment: an approach based on representation learning
Abstract Background While representation learning techniques have shown great promise in application to a number of different NLP tasks, they have had little impact on the problem of ontology matching. Unlike past work that has focused on feature engineering, we present a novel representation learni...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2018-08-01
|
| Series: | Journal of Biomedical Semantics |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s13326-018-0187-8 |
| _version_ | 1819149194329325568 |
|---|---|
| author | Prodromos Kolyvakis Alexandros Kalousis Barry Smith Dimitris Kiritsis |
| author_facet | Prodromos Kolyvakis Alexandros Kalousis Barry Smith Dimitris Kiritsis |
| author_sort | Prodromos Kolyvakis |
| collection | DOAJ |
| description | Abstract Background While representation learning techniques have shown great promise in application to a number of different NLP tasks, they have had little impact on the problem of ontology matching. Unlike past work that has focused on feature engineering, we present a novel representation learning approach that is tailored to the ontology matching task. Our approach is based on embedding ontological terms in a high-dimensional Euclidean space. This embedding is derived on the basis of a novel phrase retrofitting strategy through which semantic similarity information becomes inscribed onto fields of pre-trained word vectors. The resulting framework also incorporates a novel outlier detection mechanism based on a denoising autoencoder that is shown to improve performance. Results An ontology matching system derived using the proposed framework achieved an F-score of 94% on an alignment scenario involving the Adult Mouse Anatomical Dictionary and the Foundational Model of Anatomy ontology (FMA) as targets. This compares favorably with the best performing systems on the Ontology Alignment Evaluation Initiative anatomy challenge. We performed additional experiments on aligning FMA to NCI Thesaurus and to SNOMED CT based on a reference alignment extracted from the UMLS Metathesaurus. Our system obtained overall F-scores of 93.2% and 89.2% for these experiments, thus achieving state-of-the-art results. Conclusions Our proposed representation learning approach leverages terminological embeddings to capture semantic similarity. Our results provide evidence that the approach produces embeddings that are especially well tailored to the ontology matching task, demonstrating a novel pathway for the problem. |
| first_indexed | 2024-12-22T13:57:44Z |
| format | Article |
| id | doaj.art-55e13df2e1244b099f554976f3e41e1e |
| institution | Directory Open Access Journal |
| issn | 2041-1480 |
| language | English |
| last_indexed | 2024-12-22T13:57:44Z |
| publishDate | 2018-08-01 |
| publisher | BMC |
| record_format | Article |
| series | Journal of Biomedical Semantics |
| spelling | doaj.art-55e13df2e1244b099f554976f3e41e1e2022-12-21T18:23:31ZengBMCJournal of Biomedical Semantics2041-14802018-08-019112010.1186/s13326-018-0187-8Biomedical ontology alignment: an approach based on representation learningProdromos Kolyvakis0Alexandros Kalousis1Barry Smith2Dimitris Kiritsis3École Polytechnique Fédérale de Lausanne (EPFL)Business Informatics Department, University of Applied SciencesDepartment of Philosophy and Department of Biomedical InformaticsÉcole Polytechnique Fédérale de Lausanne (EPFL)Abstract Background While representation learning techniques have shown great promise in application to a number of different NLP tasks, they have had little impact on the problem of ontology matching. Unlike past work that has focused on feature engineering, we present a novel representation learning approach that is tailored to the ontology matching task. Our approach is based on embedding ontological terms in a high-dimensional Euclidean space. This embedding is derived on the basis of a novel phrase retrofitting strategy through which semantic similarity information becomes inscribed onto fields of pre-trained word vectors. The resulting framework also incorporates a novel outlier detection mechanism based on a denoising autoencoder that is shown to improve performance. Results An ontology matching system derived using the proposed framework achieved an F-score of 94% on an alignment scenario involving the Adult Mouse Anatomical Dictionary and the Foundational Model of Anatomy ontology (FMA) as targets. This compares favorably with the best performing systems on the Ontology Alignment Evaluation Initiative anatomy challenge. We performed additional experiments on aligning FMA to NCI Thesaurus and to SNOMED CT based on a reference alignment extracted from the UMLS Metathesaurus. Our system obtained overall F-scores of 93.2% and 89.2% for these experiments, thus achieving state-of-the-art results. Conclusions Our proposed representation learning approach leverages terminological embeddings to capture semantic similarity. Our results provide evidence that the approach produces embeddings that are especially well tailored to the ontology matching task, demonstrating a novel pathway for the problem.http://link.springer.com/article/10.1186/s13326-018-0187-8Ontology matchingSemantic similaritySentence embeddingsWord embeddingsDenoising autoencoderOutlier detection |
| spellingShingle | Prodromos Kolyvakis Alexandros Kalousis Barry Smith Dimitris Kiritsis Biomedical ontology alignment: an approach based on representation learning Journal of Biomedical Semantics Ontology matching Semantic similarity Sentence embeddings Word embeddings Denoising autoencoder Outlier detection |
| title | Biomedical ontology alignment: an approach based on representation learning |
| title_full | Biomedical ontology alignment: an approach based on representation learning |
| title_fullStr | Biomedical ontology alignment: an approach based on representation learning |
| title_full_unstemmed | Biomedical ontology alignment: an approach based on representation learning |
| title_short | Biomedical ontology alignment: an approach based on representation learning |
| title_sort | biomedical ontology alignment an approach based on representation learning |
| topic | Ontology matching Semantic similarity Sentence embeddings Word embeddings Denoising autoencoder Outlier detection |
| url | http://link.springer.com/article/10.1186/s13326-018-0187-8 |
| work_keys_str_mv | AT prodromoskolyvakis biomedicalontologyalignmentanapproachbasedonrepresentationlearning AT alexandroskalousis biomedicalontologyalignmentanapproachbasedonrepresentationlearning AT barrysmith biomedicalontologyalignmentanapproachbasedonrepresentationlearning AT dimitriskiritsis biomedicalontologyalignmentanapproachbasedonrepresentationlearning |