INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10362/142633 |
Resumo: | This study measures how Portuguese BERT performs as a semantic search system for the Portuguese Retirement Law by comparing the proposed solution answers with a test set. The study also envision edhowa feedback method (Top-Ranking)would be implemented in the developed system. This solution aimed to reorder the questions returned by the mode l based on semantic similarity. The semantic search solution showed positive results, but the Top-Ranking method failed to be implemented as the metric for semantic similarity delivered inconsistent results, thus failing to provide a solid structure for such system. The client showed very positive feedback regarding the developed tool, especially when comparing it with the current implemented solution. This project showed that Portuguese BERT is a tool with a lot of potential for semantic search on various other domains. |
id |
RCAP_36335b0117c767cc068b9b80d4c9d5b8 |
---|---|
oai_identifier_str |
oai:run.unl.pt:10362/142633 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search systemMachine learningBusiness analyticsNatural language processingChatbotNlpIncmBertHaystackLaw query systemSemantic searchFeedbackTop rankingUser experienceRetirement lawBig data analysisDomínio/Área Científica::Ciências Sociais::Economia e GestãoThis study measures how Portuguese BERT performs as a semantic search system for the Portuguese Retirement Law by comparing the proposed solution answers with a test set. The study also envision edhowa feedback method (Top-Ranking)would be implemented in the developed system. This solution aimed to reorder the questions returned by the mode l based on semantic similarity. The semantic search solution showed positive results, but the Top-Ranking method failed to be implemented as the metric for semantic similarity delivered inconsistent results, thus failing to provide a solid structure for such system. The client showed very positive feedback regarding the developed tool, especially when comparing it with the current implemented solution. This project showed that Portuguese BERT is a tool with a lot of potential for semantic search on various other domains.Xufre, PatríciaMagalhães, JoãoRUNCardoso, Diogo Nasser2022-07-29T10:30:08Z2022-01-202021-12-172022-01-20T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10362/142633TID:203022211enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2024-03-11T05:20:28Zoai:run.unl.pt:10362/142633Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-20T03:50:26.607149Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system |
title |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system |
spellingShingle |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system Cardoso, Diogo Nasser Machine learning Business analytics Natural language processing Chatbot Nlp Incm Bert Haystack Law query system Semantic search Feedback Top ranking User experience Retirement law Big data analysis Domínio/Área Científica::Ciências Sociais::Economia e Gestão |
title_short |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system |
title_full |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system |
title_fullStr |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system |
title_full_unstemmed |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system |
title_sort |
INCM & Nova Sbe PBL project - retirement law query system: using top ranking to improve user experience in a semantic search system |
author |
Cardoso, Diogo Nasser |
author_facet |
Cardoso, Diogo Nasser |
author_role |
author |
dc.contributor.none.fl_str_mv |
Xufre, Patrícia Magalhães, João RUN |
dc.contributor.author.fl_str_mv |
Cardoso, Diogo Nasser |
dc.subject.por.fl_str_mv |
Machine learning Business analytics Natural language processing Chatbot Nlp Incm Bert Haystack Law query system Semantic search Feedback Top ranking User experience Retirement law Big data analysis Domínio/Área Científica::Ciências Sociais::Economia e Gestão |
topic |
Machine learning Business analytics Natural language processing Chatbot Nlp Incm Bert Haystack Law query system Semantic search Feedback Top ranking User experience Retirement law Big data analysis Domínio/Área Científica::Ciências Sociais::Economia e Gestão |
description |
This study measures how Portuguese BERT performs as a semantic search system for the Portuguese Retirement Law by comparing the proposed solution answers with a test set. The study also envision edhowa feedback method (Top-Ranking)would be implemented in the developed system. This solution aimed to reorder the questions returned by the mode l based on semantic similarity. The semantic search solution showed positive results, but the Top-Ranking method failed to be implemented as the metric for semantic similarity delivered inconsistent results, thus failing to provide a solid structure for such system. The client showed very positive feedback regarding the developed tool, especially when comparing it with the current implemented solution. This project showed that Portuguese BERT is a tool with a lot of potential for semantic search on various other domains. |
publishDate |
2021 |
dc.date.none.fl_str_mv |
2021-12-17 2022-07-29T10:30:08Z 2022-01-20 2022-01-20T00:00:00Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10362/142633 TID:203022211 |
url |
http://hdl.handle.net/10362/142633 |
identifier_str_mv |
TID:203022211 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799138101240528896 |