1. Repositório Institucional da UFMG
  2. Publicações Científicas e Culturais
  3. Artigo de Evento
Please use this identifier to cite or link to this item: http://hdl.handle.net/1843/57494
Full metadata record
DC FieldValueLanguage
dc.creatorThiago Castro Ferreirapt_BR
dc.creatorJoão Victor de Pinho Costapt_BR
dc.creatorDaniel Hasan Dalippt_BR
dc.creatorCelso Françapt_BR
dc.creatorMarcos André Gonçalvespt_BR
dc.creatorRodrigo Bastos Fóscolopt_BR
dc.creatorAdriana Silvina Paganopt_BR
dc.creatorIsabela Rigottopt_BR
dc.creatorVitoria Portellapt_BR
dc.creatorGabriel Frotapt_BR
dc.creatorAna Luisa A. R. Guimarãespt_BR
dc.creatorAdalberto Pennapt_BR
dc.creatorIsabela Leept_BR
dc.creatorTayane A. Soarespt_BR
dc.creatorSophia Rolimpt_BR
dc.creatorRossana Cunhapt_BR
dc.creatorAriel Santospt_BR
dc.creatorRivaney F. Oliveirapt_BR
dc.creatorAbisague Langbehnpt_BR
dc.date.accessioned2023-08-04T20:03:10Z-
dc.date.available2023-08-04T20:03:10Z-
dc.date.issued2021-
dc.citation.issue2021pt_BR
dc.citation.spage234pt_BR
dc.citation.epage243pt_BR
dc.identifier.doihttps://doi.org/10.26615/978-954-452-072-4_028pt_BR
dc.identifier.isbn9789544520724pt_BR
dc.identifier.urihttp://hdl.handle.net/1843/57494-
dc.description.resumoThis study describes the development of a Portuguese Community-Question Answering benchmark in the domain of Diabetes Mellitus using a Recognizing Question Entailment (RQE) approach. Given a premise question, RQE aims to retrieve semantically similar, already answered, archived questions. We build a new Portuguese benchmark corpus with 785 pairs between premise questions and archived answered questions marked with relevance judgments by medical experts. Based on the benchmark corpus, we leveraged and evaluated several RQE approaches ranging from traditional information retrieval methods to novel large pre-trained language models and ensemble techniques using learn-to-rank approaches. Our experimental results show that a supervised transformer-based method trained with multiple languages and for multiple tasks (MUSE) outperforms the alternatives. Our results also show that ensembles of methods (stacking) as well as a traditional (light) information retrieval method (BM25) can produce competitive results. Finally, among the tested strategies, those that exploit only the question (not the answer), provide the best effectiveness-efficiency trade-off. Code is publicly available.pt_BR
dc.description.sponsorshipCNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológicopt_BR
dc.description.sponsorshipFAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Geraispt_BR
dc.description.sponsorshipCAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superiorpt_BR
dc.format.mimetypepdfpt_BR
dc.languageengpt_BR
dc.publisherUniversidade Federal de Minas Geraispt_BR
dc.publisher.countryBrasilpt_BR
dc.publisher.departmentFALE - FACULDADE DE LETRASpt_BR
dc.publisher.departmentICX - DEPARTAMENTO DE CIÊNCIA DA COMPUTAÇÃOpt_BR
dc.publisher.departmentMED - DEPARTAMENTO DE CLÍNICA MÉDICApt_BR
dc.publisher.initialsUFMGpt_BR
dc.relation.ispartofInternational Conference Recent Advances in Natural Language Processingpt_BR
dc.rightsAcesso Abertopt_BR
dc.subject.otherLingüísticapt_BR
dc.subject.otherCiência da Computaçãopt_BR
dc.subject.otherDiabetespt_BR
dc.titleEvaluating recognizing question entailment methods for a Portuguese Community Question-Answering System about Diabetes Mellituspt_BR
dc.typeArtigo de Eventopt_BR
dc.identifier.orcidhttps://orcid.org/0000-0003-0200-3646pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0002-8532-7701pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0002-3150-3503pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0002-0251-7172pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0002-2075-3363pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0001-5403-8360pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0002-8548-7625pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0003-4440-8123pt_BR
Appears in Collections:Artigo de Evento

Files in This Item:
File Description SizeFormat 
Evaluating Recognizing Question Entailment Methods for a Portuguese Community Question-Answering System about Diabetes Mellitus.pdf807.9 kBAdobe PDFView/Open
Show simple item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.