Semantic Linking of BBC Radio (SLoBR)
This project explores ways to enrich broadcasters' programme resources using Linked Data using as an example the BBC Early Music Show (EMS) and Early Music Online
The Early Music Show (EMS) is a radio programme on BBC Radio 3 which describes itself as an ‘exploration of early music. It looks at early developments in musical performance and composition, both in Britain and abroad’, playing primarily European Classical music from the eighteenth century and earlier.
As with all regular BBC programmes EMS has a dedicated area on the Corporation’s website with clips, podcasts, and supporting information about current and past editions. The BBC also exposes linked data about its programmes including EMS. Available data includes a list of featured works for each episode, and information on the contributors (performers, composers, singers, arrangers, etc.) associated with these works. Links to audio excerpts are also available for a subset of the featured works.
SLICKMEM (Crawford, Fields, Lewis & Page, 2014), a database and RDF triplestore aligning the Early Music Online (EMO) and Electronic Corpus of Lute Music (ECOLM) datasets, covers a subset of the repertory broadcasts on EMS, both in terms of historical period and genre, with EMO consisting of primarily sixteenth-century vocal music and ECOLM featuring music for the lute extending into the eighteenth century. Through the Creative-Commons licensed SLICKMEM resources 15,485 pages of digitized score representing 2,756 works by over 400 composers are available.
The Semantic Linking of BBC Radio (SLoBR) project primarily aims to enrich a visitor’s exploration of the EMS website by presenting additional information incorporated by way of data-level links with external resources. This requires us to semantically align the EMS programme data published by the BBC with data available through the other sources, in particular SLICKMEM, and via SLICKMEM with further complementary datasets from sources including DBPedia and MusicBrainz. In the process, we will develop a set of tools applicable to a wider range of alignment contexts, and an intermediary interface layer to support the creation of augmented programme interfaces for music broadcasting and beyond.
An initial approach at semantically aligning EMS and SLICKMEM has been undertaken making use of MusicBrainz IDs available for 50% of EMS contributors. Using this basic approach we were able to map 106 contributors between these datasets, related to 1,227 musical works featured on EMS and covering 185 of the 499 episodes currently available (37%). We hope to extend our coverage beyond this initial set of alignments by employing fuzzy matching techniques on additional parts of the data (e.g. composer names, work titles).