Linked data – the machine readable web, as the inventor of the Web Tim Berners Lee explains here is the next layer of the web. He developed a five star rating to describe information which is fully compliant with the semantic web and allows it to become part of the growing web of linked data. In the recent discussion paper IKMemergent explained why this was relevant for the international development community.
As a result of participation in the IKMemergent Workshop in Oxford, UK, in November 2010, IFPRI has taken the Global Hunger Index (GHI) as an example and published it as a linked data RDF files and documented the experience. We worked with Practical participation to develop the initial files and seek guidance on the options and approaches for publishing linked data. This is the first stage of the project to make the data available and then monitor its use and look at ways to promote and integrate it with more datasets.
Groups have already used the original data of the GHI to produce new mappings of the data, see the examples from Tableau and Chartsbin. By publishing the linked data we hope others will integrate the information with other visualizations and disseminate the results of the report.
Linked data has already been used by the international community to bring together different datasets, for example comparing Aid from the UK and USA.
The process of initially setting up a process to publish linked data sets is involved, but obviously with a recurring dataset such as the Global hunger Index it is easy to update.
Although the datasets prove challenging to convert, information on many of the organisations outputs is already available in Linked data form. The hundred or so wikipedia pages featuring links and descriptions of IFPRI output are already part of the web of linked data through Dbpedia. A sample of metadata describing IFPRI presentations, collections and twitter accounts can be seen through Sindice.
Get the data
At present the GHI 2010 data is made available in two different forms and is available in raw RDF/XML and N3 files. http://data.ifpri.org/rdf/ghi/
|Chris Addison|Head of Web Communication IFPRI
Filed under: Uncategorized