Skip to Main content Skip to Navigation
Conference papers

Expérimentations autour des architectures d'apprentissage par transfert pour l'extraction de relations biomédicales

Walid Hafiane 1 Joël Legrand 2 Yannick Toussaint 1 Adrien Coulet 1, 3, 4
1 ORPAILLEUR - Knowledge representation, reasonning
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
2 SYNALP - Natural Language Processing : representations, inference and semantics
LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Relation extraction (RE) consists in identifying and structuring automatically relations of interest from texts. Recently, BERT improved the top performances for several NLP tasks, including RE. However, the best way to use BERT, within a machine learning architecture, and within a transfer learning strategy is still an open question since it is highly dependent on each specific task and domain. Here, we explore various BERT-based architectures and transfer learning strategies (i.e., frozen or fine-tuned) for the task of biomedical RE on two corpora. Among tested architectures and strategies, our *BERT-segMCNN with fine-tuning reaches performances higher than the state-of-the-art on the two corpora (1.73 % and 32.77 % absolute improvement on ChemProt and PGxCorpus corpora respectively). More generally, our experiments illustrate the expected interest of fine-tuning with BERT, but also the unexplored advantage of using structural information (with sentence segmentation), in addition to the context classically leveraged by BERT.
Complete list of metadatas

https://hal.inria.fr/hal-03073601
Contributor : Adrien Coulet <>
Submitted on : Wednesday, December 16, 2020 - 11:23:58 AM
Last modification on : Tuesday, January 19, 2021 - 3:32:41 AM

File

hafiane_et_al_fr.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03073601, version 1

Citation

Walid Hafiane, Joël Legrand, Yannick Toussaint, Adrien Coulet. Expérimentations autour des architectures d'apprentissage par transfert pour l'extraction de relations biomédicales. EGC2021 - Extraction et Gestion des Connaissances (EGC), Jan 2021, Montpellier/Virtuel, France. ⟨hal-03073601⟩

Share

Metrics

Record views

36

Files downloads

88