Publicación:
A Novel Methodology for Enhancing Cross-language and Domain Adaptability in Temporal Expression Normalization

Vista sencilla de ítem
dc.contributor.authorSánchez de Castro Fernández, Alejandro
dc.contributor.authorAraujo Serna, M. Lourdes
dc.contributor.authorMartínez Romo, Juan
dc.contributor.funderAgencia Estatal de Investigación (España)
dc.contributor.funderUniversidad Nacional de Educación a Distancia (UNED)
dc.date.accessioned2026-01-27T08:50:00Z
dc.date.available2026-01-27T08:50:00Z
dc.date.issued2025-04-19
dc.descriptionThe registered version of this article, first published in “Computational Linguistics 51 (4): 1303–1335", is available online at the publisher's website: https://doi.org/10.1162/COLI.a.12
dc.descriptionLa versión registrada de este artículo, publicado por primera vez en “Computational Linguistics 51 (4): 1303–1335", está disponible en línea en el sitio web del editor: https://doi.org/10.1162/COLI.a.12
dc.description.abstractAccurate temporal expression normalization, the process of assigning a numerical value to a temporal expression, is essential for tasks such as timeline creation and temporal reasoning. While rule-based normalization systems are limited in adaptability across different domains and languages, deep-learning solutions in this area have not been extensively explored. An additional challenge is the scarcity of manually annotated corpora with temporal annotations. To address the adaptability limitations of current systems, we propose a highly adaptable methodology that can be applied to multiple domains and languages. This can be achieved by leveraging a multilingual Pre-trained Language Model (PTLM) with a fill-mask architecture, using a Value Intermediate Representation (VIR) where the temporal expression value format is adjusted to the fill-mask representation. Our approach involves a two-phase training process. Initially, the model is trained with a novel masking policy on a large English biomedical corpus that is automatically annotated with normalized temporal expressions, along with a complementary hand-crafted temporal expressions corpus. This addresses the lack of manually annotated data and helps to achieve sufficient capacity for adaptation to diverse domains or languages. In the second phase, we show how the model can be tailored to different domains and languages using various techniques, showcasing the versatility of the proposed methodology. This approach significantly outperforms existing systems.en
dc.description.provenanceMade available in DSpace on 2026-01-27T08:50:00Z (GMT). No. of bitstreams: 1 MartinezRomo_Juan_TemporalExpression_JUAN MARTÍNEZ ROMO.pdf: 1222597 bytes, checksum: aee350684f77feb1938cf6ad4abc9c99 (MD5) Previous issue date: 2025-04-19en
dc.description.sponsorshipThis work has been funded by the following projects: OBSER-MENH (MCIN/AEI/10.13039/501100011033 and NextGenerationEU”/PRTR with identification TED2021-130398B-C21), SICAMESP (2023-VICE-0029), and by the project EDHER-MED (PID2022-136522OB-C21)”.
dc.description.versionversión original
dc.identifier.citationSánchez de Castro, A., Araujo, L., & Martinez-Romo, J. (2025). A Novel Methodology for Enhancing Cross-Language and Domain Adaptability in Temporal Expression Normalization. Computational Linguistics 51 (4): 1303–1335. https://doi.org/10.1162/COLI.a.12
dc.identifier.doihttps://doi.org/10.1162/COLI.a.12
dc.identifier.issn0891-2017
dc.identifier.urihttps://hdl.handle.net/20.500.14468/31578
dc.journal.issue4
dc.journal.titleComputational Linguistics
dc.journal.volume51
dc.language.isoen
dc.page.final1335
dc.page.initial1303
dc.publisherMassachusetts Institute of Technology Press
dc.relation.centerE.T.S. de Ingeniería Informática
dc.relation.departmentLenguajes y Sistemas Informáticos
dc.relation.projectidinfo:eu-repo/grantAgreement/AEIProyectos Estratégicos Orientados a la Transición Ecológica y a la Transición Digital 2021/TED2021-130398B-C21/ES/GELP: Generación mediante procesamiento del lenguaje de perfiles demográficos en redes sociales para la detección de riesgo de suicidio y su relación con otros problemas psicológicos
dc.relation.projectidinfo:eu-repo/grantAgreement/AEI/Proyectos de I+D+I (Generación de Conocimiento y Retos Investigación) 2022/PID2022-136522OB-C21/ES/Detección precoz de enfermedades de alto impacto mediante el procesamiento del lenguaje natural
dc.rightsinfo:eu-repo/semantics/openAccess
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/deed.es
dc.subject1203 Ciencia de los ordenadores
dc.titleA Novel Methodology for Enhancing Cross-language and Domain Adaptability in Temporal Expression Normalizationen
dc.typeartículoes
dc.typejournal articleen
dspace.entity.typePublication
relation.isAuthorOfPublication6c2a060d-18a1-4a12-b8d1-d41a6d335ec7
relation.isAuthorOfPublication77c4023e-4374-442a-9dfb-b9d4b609c31e
relation.isAuthorOfPublication91b7e317-2a30-494f-98e9-3a0e026747b1
relation.isAuthorOfPublication.latestForDiscovery6c2a060d-18a1-4a12-b8d1-d41a6d335ec7
Archivos
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
MartinezRomo_Juan_TemporalExpression_JUAN MARTÍNEZ ROMO.pdf
Tamaño:
1.17 MB
Formato:
Adobe Portable Document Format
Descargar
Bloque de licencias
Mostrando 1 - 1 de 1
No hay miniatura disponible
Nombre:
license.txt
Tamaño:
3.62 KB
Formato:
Item-specific license agreed to upon submission
Descripción:
Descargar
Colecciones
Artículos y papers