BBC R&D, London, UK
The BBC generates a vast quantity of information, broadcasting between 1,000 and 1,500 programmes per day and generating a significant news output. The BBC has also accumulated a very large archive of TV and radio programmes, as well as pictures, texts, musical scores etc. since 1922. Managing all this information constitutes a major challenge.
In this talk, we are going to describe some recent work undertaken by BBC R&D to investigate how a mixture of Semantic Web technologies, machine-generated annotations and crowd-sourcing can enable us to publish large archives of content very quickly.
Yves Raimond holds a PhD from Queen Mary, University of London. His thesis was entitled ‘A Distributed Music Information System’, and defined a framework for applying a range of Semantic Web technologies for managing and distributing music-related information. As part of his thesis, he contributed extensively to what would become the ‘Linking Open Data’ community project. Since 2008, he has been working for the BBC, first on the bbc.co.uk/programmes service, publishing structured data about all BBC programmes, and then in BBC R&D on unlocking archives by automatically interlinking them with related datasets. As part of this project he has worked on a prototype combining automated interlinking with Linked Data sources and crowdsourcing to open up the BBC World Service archive.