Measuring the effect of different types of unsupervised word representations on Medical Named Entity Recognition.

Journal: International journal of medical informatics
Published Date: Sep 1, 2019

Abstract

BACKGROUND: This work deals with Natural Language Processing applied to the clinical domain. Specifically, the work deals with a Medical Entity Recognition (MER) on Electronic Health Records (EHRs). Developing a MER system entailed heavy data preprocessing and feature engineering until Deep Neural Networks (DNNs) emerged. However, the quality of the word representations in terms of embedded layers is still an important issue for the inference of the DNNs.

Authors

  • Arantza Casillas
    IXA Group, University of the Basque Country (UPV-EHU), Computer Engineering Faculty, P. Manuel Lardizabal, 1, 20018 Donostia-San Sebastián, Spain(1). Electronic address: arantza.casillas@ehu.eus.
  • Nerea Ezeiza
    IXA Group, University of the Basque Country (UPV-EHU), Manuel Lardizabal 1, 20080 Donostia, Spain. Electronic address: n.ezeiza@ehu.eus.
  • Iakes Goenaga
    IXA Group, University of the Basque Country (UPV-EHU), Manuel Lardizabal 1, 20080 Donostia, Spain. Electronic address: iakes.goenaga@ehu.eus.
  • Alicia Pérez
    IXA Group, University of the Basque Country (UPV-EHU), Computer Engineering Faculty, P. Manuel Lardizabal, 1, 20018 Donostia-San Sebastián, Spain(1).
  • Xabier Soto
    IXA Group, University of the Basque Country (UPV-EHU), Manuel Lardizabal 1, 20080 Donostia, Spain. Electronic address: xabier.soto@ehu.eus.

Keywords

External Resources

Popular Topics
Recent Journals