Put your hands up if you haven’t heard of the BBC! None of you? Thought so… Not only will this talk give you an insight into the unique challenges and opportunities alike the BBC faces with data, but you will also gain an entry-level understanding Semantic Web technologies, and how concepts such as knowledge graphs can provide powerful ways to analyse and emphasise relationships in data.
As a public service organisation, the BBC strives to achieve its mission to inform, educate and entertain. Without the intelligent use of data, this would be a challenging affair. Thankfully, we at the BBC have a little bit of data… Okay, we have a lot. How should we use it? Are technologies of the Semantic Web, specifically Linked Data, the answer? I will be posing the challenges we face in this section, talking through the unique aspects that arise when working with data as a public service organisation, and perhaps I’ll tell you a little about myself.
The Semantic Web, proposed by Tim Berners-Lee, extends the standards of the World Wide Web to promote common data formats and exchange protocols; Linked Data is a progression of these ideas, focused on a more practical implementation. During this part of the talk, concepts that constitute the Linked Data will be picked apart and explained, with no expectations of prior knowledge. These concepts include, but are not limited to, the Resource Description Framework (RDF), the Web Ontology Language (OWL), the SPARQL query language, and, at a higher level, what a knowledge graph is. Limitations will discussed, such as the implications of RDF only supporting binary predicates, and comparisons will be drawn to other data storage paradigms. On finishing this section, you will have an idea of the concepts that the Semantic Web depends on, and even better, you will know when to use them.
The concepts discussed so far will be applied here using the python library
RDFLib, among other tools. This will leave you with the skills to get started on your own RDF-fueled projects.
In the past, the BBC have used Linked Data technologies in a number of ways, including facilitation of outstanding Olympics coverage and powering News and Sport behind the scenes. Now, there is great incentive to create pan-BBC information retrieval systems with a focus on semantics, being pushed by recently developed in-house systems for automated content tagging. Talking through applications of these technologies in a wider context will give you a better idea of the power of them at scale, the ethics, and the associated responsibilities we have as a public service organisation.