Blind links, a big challenge in the linked data idea: Analysis of Persian Subject Headings

• Published in: Pizhūhishnāmah-i pardāzish va mudiriyyat-i iṭṭilāʻāt (Online) Vol. 30; no. 1; pp. 223 - 244
• Main Author: Atefeh Sharif
• Format: Journal Article
• Language: Persian
• Published: Iranian Research Institute for Information and Technology 01.12.2014
• Subjects: Library of Congress Subject Headings (LCSHs); linked data; Nosa books; Persian Subject Headings (PSHs); semantic web
• ISSN: 2251-8223; 2251-8231

In this survey, Linked data concept as exposing, sharing, and connecting pieces of data, information, and knowledge on the Semantic Web and some potential problems in converting Persian subject headings (PSHs) Records into linked data were discussed. A data set (11233 records) of PSHs was searched in three information retrieval systems including National Library of Iran (NLI) online catalog, Library of Congress (LC) online catalog and NOSA books. Correct links between Persian and English subject headings in the 9519 common records of two catalogs were recorded. The results indicate that the links between Persian and English subjects in 20% of records were failed. The maximum error was associated with the anonymous databases (6/7 %) in NLI online catalog. It is recommended to preprocess the PSHs records before any conversion projects. It seems that, during the preprocessing, the potential errors could be identified and corrected.