Projects in the area of web access

BnF is involved in projects in partnership with other national and international institutions in the area of access to information via the web. The data currently available in BnF’s catalogs, which are created daily by professionals, serve as a platform for all these projects.
Certain search engines can now be used to carry out statistical or association-based analysis (based on the frequency of words in documents), linguistic analysis (on the use of vocabulary and grammar) and semantic analysis (automatic classification of the words in a text, navigation in a thesaurus, etc.), drawing in particular on controlled vocabularies.

Multilingual access to subjects

MACS = Multilingual ACcess to Subjects

The MACS program, launched by the CENL in 1997, brings together the Bibliothèque nationale de France, the Deutsche Nationalbibliothek and the Swiss National Library. Its objective is to develop multilingual access to subjects, based on three indexing languages, RAMEAU for French, LCSH for English and SWD for German.

Semantic web and interoperability of vocabulary

Experimenting in the area of illuminations

This experiment was conducted as part of STITCH, a European research project aimed at finding solutions to facilitate navigation between heritage collections with heterogeneous metadata.

Like MACS, STITCH explores the semantic branch of the research and addresses the challenge of multilingualism and the use of multiple vocabularies in subject indexing. However, while MACS seeks to manually identify correlations between terms in more than one language, STITCH studies the use of automated techniques from the semantic web, including in particular ontology alignment.

The experiment covered the alignment of two vocabularies used to index illuminations in the Mandragore database – the BnF Manuscripts Department’s iconographic database – and the Illuminated Medieval Manuscripts of the Netherlands National Library (KB). To enable alignment, each of the two vocabularies was first converted into SKOS.

The aim of the experiment was to demonstrate that techniques from the semantic web can be used to carry out simultaneous semantic searches across more than one digital heritage collection. The experiment led to the creation of a prototype based on a sample of the original collections. The prototype is accessible online and can be searched in various ways.

Semantic web

The semantic web refers to a range of technologies used to create a detailed and structured description of web data with the aim of providing efficient and comprehensive access to those data. In particular, the semantic web is based on the RDF (Resource Description Framework) model, intended to describe data and the relationships between them. The benefit of the semantic web for libraries lies in the ability to create more links between catalog data, in a similar way to data models such as FRBR, and to increase the visibility of those data on the web.

Friday, March 23, 2012