Linked Data for Open and Distance Learning – Part 1
Continued from Linked Data for Open and Distance Learning – Part 0…
What is Linked Data
The foundation of the Web is that it is a network of documents connected by hyperlinks (see Figure 1a). Each document is identified by a Web address, a URI (Berners-Lee et al, 1998), and might represent a document which content is encoded using a standard, universally readable format (most commonly HTML ). Following this, the simplest way to describe Linked Data is that it is about using these same principles of the Web architecture not only for documents, but also for data.
The foundation of Linked Data is therefore that data objects on the Web are identified, similarly to documents, by URIs. The representation of the data—i.e. the information associated with a data object— is then represented by Web links, which can themselves be characterised by URIs. This makes it possible to represent information in such a way that it is materialised as a graph, where nodes are URIs or literal data values (strings, numbers) and the edges are links between them (see Figure 1b). For example, a university like The Open University publishes information about the courses it offers through its website, as well as using linked data . It achieves that by assigning to every course a dedicated URI that acts both as an identifier for the course on the Web, and as a way to access structured information about this course. For example, http://data.open.ac.uk/course/aa100 is the URI for the course with code AA100, which is an undergraduate (level 1) course in Arts and Humanities, entitled “The arts past and present”. Through the links between this URI and others, information about this course is being represented regarding the topics and description of the course, where it is available, how it is assessed, what course material and open educational resources relate to it, etc. (see Figure 2). While most of the other data objects it relates to are also identified by URIs within the domain of the Open University, it is important to remark here that it links to other data sources, such as the UK government’s information about the Open University or information provided by the Geonames platform about the countries in which the course is available. This demonstrates how, from these basic principles, information originating from widely different systems and sources can be seamlessly integrated.
While linked data has, for a few years, been confined in the corresponding academic research community, it is now gaining significant momentum in many different areas (see Figure 3 showing a catalog of interlinked datasets). Governments (most notably in the US and the UK ) are in particular leading open data initiatives where information about aspects such as transport, environment, public spending and education are being made available as Linked Data, to be easily referenced, accessed and reused by the public. Datasets, such as the one of the geonames initiative are also being made available as Linked Data, making it possible for anybody to link to reference information regarding geographical places in the world. One of the most reused source of information on this Web of Data is DBpedia , a Linked Data version of Wikipedia providing easily reusable and query-able general information in almost every domain. DBpedia basically transforms every page in Wikipedia into a linked data entity, analysing the “infobox” on the Wikipedia page to extract structured information about each entity, and their relationships to others. Many cultural organisations such as museums and libraries have also realized the potential of the linked data, and started publishing open data, including for example the content of their catalogues and collections, as Linked Data. Finally, Linked Data is more and more used by universities and other education institutions to make their public information and open resources more accessible, discoverable and reusable (this report includes several examples in this area).
A REPORT ON: Linked Data for Open and Distance Learning
Second Edition July 2014
Prepared for the Commonwealth of Learning
by: Mathieu d’Aquin Research Fellow Knowledge Media Institute The Open University, UK