English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Relation extraction (RE) consists in identifying and structuringautomatically relations of interest from texts. Recently, BERT improved the topperformances for several NLP tasks, including RE. However, the best way to useBERT, within a machine learning architecture, and within a transfer learningstrategy is still an open question since it is highly dependent on eachspecific task and domain. Here, we explore various BERT-based architectures andtransfer learning strategies (i.e., frozen or fine-tuned) for the task ofbiomedical RE on two corpora. Among tested architectures and strategies, our*BERT-segMCNN with finetuning reaches performances higher than thestate-of-the-art on the two corpora (1.73 % and 32.77 % absolute improvement onChemProt and PGxCorpus corpora respectively). More generally, our experimentsillustrate the expected interest of fine-tuning with BERT, but also theunexplored advantage of using structural information (with sentencesegmentation), in addition to the context classically leveraged by BERT.

Experiments on transfer learning architectures for biomedical relation extraction