Biomedical ontology alignment: an approach based on representation learning.

Journal: Journal of biomedical semantics
Published Date:

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.

Authors

  • Prodromos Kolyvakis
    École Polytechnique Fédérale de Lausanne (EPFL), Route Cantonale, Lausanne, 1015, Switzerland. prodromos.kolyvakis@epfl.ch.
  • Alexandros Kalousis
    Business Informatics Department, University of Applied Sciences, HES-SO, Western Switzerland Carouge, Switzerland.
  • Barry Smith
    Department of Philosophy, University at Buffalo, NY, USA.
  • Dimitris Kiritsis
    École Polytechnique Fédérale de Lausanne (EPFL), Route Cantonale, Lausanne, 1015, Switzerland.