Neural architecture for temporal relation extraction: A Bi-LSTM approach for detecting narrative containers

J. Tourille (1) , Olivier Ferret (2) , X. Tannier (1) , Aurélie Névéol (1)

1 LIMSI - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur
2 DIASI - Département Intelligence Ambiante et Systèmes Interactifs

We present a neural architecture for containment relation identification between medical events and/or temporal expressions. We experiment on a corpus of de-identified clinical notes in English from the Mayo Clinic, namely the THYME corpus. Our model achieves an F-measure of 0.613 and outperforms the best result reported on this corpus to date.

Keywords: Clinical notes; Computational linguistics; F measure; Neural architectures; Temporal relation; Temporal expressions; Linguistics

J. Tourille, Olivier Ferret, X. Tannier, Aurélie Névéol. Neural architecture for temporal relation extraction: A Bi-LSTM approach for detecting narrative containers. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, Jul 2017, Vancouver, Canada. pp.224-230, ⟨10.18653/v1/P17-2035⟩. ⟨cea-01841667⟩