Finding Knowledge - Linking Open Educational Resources
I’ve mentioned Open Data and Linked Data in my last couple of posts. Brilliantly enough my department does a 'forum meeting' once a quater, and in last week's I heard about a project (which KMi is a partner on) on just this topic!
Linking Open Educational Resources (OER)
The LinkedUp project is aiming to draw on web data – from OER metadata, to linked data – to provide university degree-level education. Stefan Dietz said a bit more about this in a talk ‘Linked Data and OER – towards a symbiotic relationship’ - which I'll talk about a bit more below.
But first, if you're not familiar with Linked Data, there’s a great indtroductory video below which explains the idea brilliantly with examples.
If you can't watch that, the idea is basically to ensure that when we put things on the web, we provide them with identifiers (normally these are web addresses), and that we provide useful information about the thing alongside it - in education, this might include what sort of file it is, what content it covers, what age group; and that it expresses relationships to other data - so to continue the example, it might express the relationship between course modules (this one first, then that one), or cross-over material, and so on. The idea is to do this, in such a way as to make the materials discoverable, using standardised formats to allow machines to select, and relate objects on the web.
Why would we link up OER?
Within the LinkedUp project, as I understand, the desire is to get organisations with OER (OpenCourseWare, OpenLearn, Merlot, etc.) with structured APIs which can be easily co-analysed, and – with a mixture of manual and automated work – semantic content (and ambiguity) assigned by looking at what appears in the API. Thus, we know HPV refers to the virus by the content it occurs alongside (as opposed to High Production Volume, or whatever). This allows us to annotate the resource with that information and look for similar resources across the set of resources (in various different databases).
If that can be accomplished, some examples of outcomes would be:
- We could search across the web for OER
- We could structure OER in terms of concept connectedness (the ‘Knowledge Graph’)
- These first two applications could be used to implement game based learning and auto-generate assessments
- They could also be used for the enrichment of learning resources (e.g. by automatically ‘bringing in’ associated multimedia resources)
Linked, Personalised, OER
Even better, in another talk (p.60 onwards) Hendrik Drachsler talks about using this data alongside personal data for true recommender systems in education. So not only would we be able to join up OER across the web, we’d be able to represent your exposure to those OER and ‘plug’ the gaps (perhaps using games and dynamic assessments).
This has obvious implications for learners, if we can facilitate meaningful access to high quality OER resources, especially interactive and engaging ones (rather than just raw text). It also has implications for how we think about education – both within educational institutions, including universities and schools, but also within other institutions who engage in educational work (museums, health organisations, charities, etc.) and of course, informal learning settings – particularly the desire to build a world of ‘lifelong’ learners, who are constantly engaged in (and with) learning for its own sake.
What do people see as the potential for this in their area?
What are the challenges (especially non-technological ones)?
Are there any good examples at the moment?
Comments below the line or @sjgknight