Training and intrinsic evaluation of lightweight word embeddings for the clinical domain in Spanish

Simple item page
dc.contributor.authorChiu, Carolina
dc.contributor.authorVillena, Fabián
dc.contributor.authorMartin, Kinan
dc.contributor.authorNúñez, Fredy R.
dc.contributor.authorBesa Correa, Cecilia
dc.contributor.authorDunstan, Jocelyn
dc.date.accessioned2022-12-28T15:35:05Z
dc.date.available2022-12-28T15:35:05Z
dc.date.issued2022
dc.description.abstractResources for Natural Language Processing (NLP) are less numerous for languages different from English. In the clinical domain, where these resources are vital for obtaining new knowledge about human health and diseases, creating new resources for the Spanish language is imperative. One of the most common approaches in NLP is word embeddings, which are dense vector representations of a word, considering the word's context. This vector representation is usually the first step in various NLP tasks, such as text classification or information extraction. Therefore, in order to enrich Spanish language NLP tools, we built a Spanish clinical corpus from waiting list diagnostic suspicions, a biomedical corpus from medical journals, and term sequences sampled from the Unified Medical Language System (UMLS). These three corpora can be used to compute word embeddings models from scratch using Word2vec and fastText algorithms. Furthermore, to validate the quality of the calculated embeddings, we adapted several evaluation datasets in English, including some tests that have not been used in Spanish to the best of our knowledge. These translations were validated by two bilingual clinicians following an ad hoc validation standard for the translation. Even though contextualized word embeddings nowadays receive enormous attention, their calculation and deployment require specialized hardware and giant training corpora. Our static embeddings can be used in clinical applications with limited computational resources. The validation of the intrinsic test we present here can help groups working on static and contextualized word embeddings. We are releasing the training corpus and the embeddings within this publication.
dc.fechaingreso.objetodigital2022-12-28
dc.fuente.origenSIPA
dc.identifier.doi10.3389/frai.2022.970517
dc.identifier.urihttps://www.frontiersin.org/articles/10.3389/frai.2022.970517/full
dc.identifier.urihttps://repositorio.uc.cl/handle/11534/66147
dc.information.autorucFacultad de letras; Núñez, Fredy R.; 0000-0002-0643-6628; 157277
dc.information.autorucEscuela de medicina; Besa Correa, Cecilia; 0000-0002-0015-0434; 167343
dc.language.isoen
dc.nota.accesoContenido completo
dc.revistaFrontiers in Artificial Intelligence
dc.rightsacceso restringido
dc.subjectNatural language processing
dc.subjectSpanish language
dc.subjectWord embeddings
dc.subjectMedical informatics
dc.subjectNeural networks
dc.subjectIntrinsic evaluation
dc.subjectSemantic evaluation
dc.titleTraining and intrinsic evaluation of lightweight word embeddings for the clinical domain in Spanish
dc.typeartículo
dc.volumen5
sipa.codpersvinculados157277
sipa.codpersvinculados167343
sipa.indexScopus
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
frai-05-970517.pdf
Size:
333.38 KB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Artículos de revistas