Metraipolis › archive › dc_fb0831fa33fe

A pre-training and self-training approach for biomedical named entity recognition — Europe PMC PMID:33561139

Source: Europe PMC (en)
Original: https://europepmc.org/article/MED/33561139
DOI: 10.1371/journal.pone.0246310
License (as archived): CC0-1.0
Archived: 2026-11-06T00:32:08.366Z
SHA-256 of the archived text: c537b39844c9371ef5904100f5058668737f3a29015fb4eee5f84e0d961cd176
Stable URL of this page: https://metraipolis.com/documento/dc_fb0831fa33fe

How to cite

A pre-training and self-training approach for biomedical named entity recognition — Europe PMC PMID:33561139 — Europe PMC (en), https://europepmc.org/article/MED/33561139 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fb0831fa33fe (Metraipolis archive dc_fb0831fa33fe, sha256 c537b39844c9371ef5904100f5058668737f3a29015fb4eee5f84e0d961cd176)

BibTeX

@misc{dcfb0831fa33fe,
  title = {A pre-training and self-training approach for biomedical named entity recognition — Europe PMC PMID:33561139},
  howpublished = {Europe PMC (en)},
  url = {https://europepmc.org/article/MED/33561139},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fb0831fa33fe; sha256 c537b39844c9371ef5904100f5058668737f3a29015fb4eee5f84e0d961cd176},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: Europe PMC (en), https://europepmc.org/article/MED/33561139. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Named entity recognition (NER) is a key component of many scientific literature mining tasks, such as information retrieval, information extraction, and question answering; however, many modern approaches require large amounts of labeled training data in order to be effective. This severely limits the effectiveness of NER models in applications where expert annotations are difficult and expensive to obtain. In this work, we explore the effectiveness of transfer learning and semi-supervised self-training to improve the performance of NER models in biomedical settings with very limited labeled data (250-2000 labeled samples). We first pre-train a BiLSTM-CRF and a BERT model on a very large general biomedical NER corpus such as MedMentions or Semantic Medline, and then we fine-tune the model on a more specific target NER task that has very limited training data; finally, we apply semi-supervised self-training using unlabeled data to further boost model performance. We show that in NER tasks that focus on common biomedical entity types such as those in the Unified Medical Language System (UMLS), combining transfer learning with self-training enables a NER model such as a BiLSTM-CRF or BERT to obtain similar performance with the same model trained on 3x-8x the amount of labeled data. We further show that our approach can also boost performance in a low-resource application where entities types are more rare and not specifically covered in UMLS.

Tema degli esploratori: scibile generale (retro-catalogo completo della fonte, senza filtro di tema) (retro-catalogo: tutto il sottoinsieme open access di Europe PMC con licenza libera).
Pubblicazione: PloS one, 2021 (prima pubblicazione 2021-02-09).
Identificativi: PMID 33561139 · PMCID PMC7872256 · DOI 10.1371/journal.pone.0246310.
Descrittori MeSH principali: Artificial Intelligence; Unsupervised Machine Learning; Recognition, Psychology.
Pagina: https://europepmc.org/article/MED/33561139 — https://doi.org/10.1371/journal.pone.0246310. Autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza dell'articolo dichiarata da Europe PMC: CC0 (la versione è indicata dall'editore sulla pagina). Abstract riportato senza modifiche al contenuto, con attribuzione; deposito marcato CC0.
Letto il 2026-11-06T00:28:36.836Z tramite l'API REST ufficiale di Europe PMC (EMBL-EBI), nel rispetto del robots.txt e del crawl-delay.