Utilize este identificador para referenciar este registo:
https://hdl.handle.net/1822/74782
Título: | Towards Entity Linking, NER in archival finding aids |
Autor(es): | Cunha, Luís Filipe da Costa Ramalho, José Carlos |
Palavras-chave: | Archival Descriptions Named Entity Recognition Machine Learning Web |
Data: | 13-Set-2021 |
Editora: | CEUR-Ws |
Revista: | CEUR Workshop Proceedings |
Resumo(s): | The amount of information present in Portuguese archives has been increasing exponentially over the years. At the moment, the majority of the data is already available to the public in digital format, however, the records are stored as unstructured text, making its data processing challenging. In this way, it is intended to perform a semantic interpretation of these documents through the identification and classification of Named Entities. For this purpose, the use of Natural Language Processing tools is proposed, training Machine Learning algorithms capable of accurately recognizing entities in this context. Finally, it is presented a Web platform that implements all the models trained in this paper, as well as some tools that gave support to the entity extraction process. |
Tipo: | Artigo em ata de conferência |
URI: | https://hdl.handle.net/1822/74782 |
ISSN: | 1613-0073 |
Versão da editora: | http://ceur-ws.org/Vol-3019/LinkedArchives_2021_paper_12.pdf |
Arbitragem científica: | yes |
Acesso: | Acesso aberto |
Aparece nas coleções: |
Ficheiros deste registo:
Ficheiro | Descrição | Tamanho | Formato | |
---|---|---|---|---|
LinkedArchives_2021_paper_12.pdf | artigo | 448,89 kB | Adobe PDF | Ver/Abrir |
Este trabalho está licenciado sob uma Licença Creative Commons