Evaluation of Mappings from MARC to Linked Data

Hyoungjoo Park, Margaret E.I. Kipp

Abstract


The purpose of this study is to assess the quality and compatibility of library linked data (LLD) schemas in use or proposed for library resources. Linked Data (LD) has the potential to provide high quality metadata on the web with the ability to incorporate existing structured data from MARC via a mapping.  Researchers selected representative libraries such as Harvard University Library, LC BIBFRAME (Library of Congress Bibliographic Framework), OCLC (Online Computer Library Canter) WorldCat, and National Library of Spain. For LD frameworks, four resources are matched into specific categories with MARC (MAchine-Readable Cataloging) tags so that it could be retrieved in both OCLC LD and BIBFRAME with the conversion tool at bibframe.org: (1) Classic, ebook,
and fiction, (2) multiple authors and part of a series, and non-fiction, (3) varying title, translation, and fiction, and (4) sub title, non-fiction. This study revealed that the choices and elements of each library made in local decisions might bring interoperability issues for LD services due to the quality metadata creation issues.


Full Text:

PDF


DOI: http://dx.doi.org/10.7152/acro.v25i1.14908