A conversão de registros na implantação de repositórios institucionais: o caso do Repositório Institucional UNESP

Carregando...
Imagem de Miniatura

Data

2014-11-21

Autores

Assumpção, Fabrício Silva [UNESP]
Silva, Renata Eleuterio da [UNESP]
Ferreira, Jaider Andrade [UNESP]
Bastos, Flávia Maria [UNESP]

Título da Revista

ISSN da Revista

Título de Volume

Editor

Universidade Federal de Minas Gerais (UFMG)

Resumo

O Repositório Institucional UNESP foi criado em 2013 e, para sua implantação, foi povoado com dados obtidos de forma automática. Considerando a experiência realizada na UNESP, este trabalho tem por objetivo apresentar os processos utilizados para a conversão dos registros coletados de três diferentes fontes de dados (Web of Science, SciELO e Scopus) para inclusão no repositório. A partir da coleta dos registros, os padrões de metadados da Web of Science, da SciELO e da Scopus foram mapeados para o perfil de aplicação de metadados utilizado no repositório. Os registros foram coletados como arquivos XML e, para sua conversão, foram elaboradas folhas de estilo utilizando a linguagem XSLT. Após essa conversão, os arquivos XML foram convertidos em arquivos CSV e, então, importados no Repositório. Conclui-se que os processos de conversão utilizados permitiram alcançar as metas iniciais do Repositório e evitaram a necessidade de inclusão dos registros de forma manual.
The UNESP Institutional Repository was started in 2013 and, for its implementation, it was populated with data harvested automatically from three distinct sources (Web of Science, SciELO and Scopus). Based on UNESP’s case, this paper aims to present the process used in the conversion of records from three distinct sources (Web of Science, SciELO and Scopus) for inclusion into the repository. After the harvesting, the Web of Science, the SciELO and the Scopus metadata standards were mapped to the metadata application profile used in the UNESP Institutional Repository. Since records were harvested as XML files, we created stylesheets using XSLT to transform these files in XML files in accordance with DSpace markup language and the application profile. From this point, the XML files were converted to CSV files in order to be imported into the Repository. We conclude that the conversion process allowed us to achieve the initial goals of Repository and to avoid the manual entering of records.

Descrição

Palavras-chave

Repositório institucional, Conversão de registros, Folha de estilos XSLT, DSpace, Padrão de metadados, Repositório Institucional UNESP, Institutional repository, Record conversion, XSLT stylesheet, Metadata standards

Como citar

In: SEMINÁRIO NACIONAL DE BIBLIOTECAS UNIVERSITÁRIAS, 18., 2014, Belo Horizonte. Anais... Belo Horizonte: UFMG, 2014. p. 1-16.