Week 4: Smarter curriculum: semantic web, linked data, and learning content

Date: February 13 - February 19


A short video reviewing the course activity to date and detailing where we're going next is available here: http://blip.tv/gsiemens/lak12-week-4-intro-video-5952769

Over the first three weeks of LAK12, we explored the context that has generated interest in learning analytics, the different flavours of analytics (business intelligence, educational data mining, learning analytics, and academic analytics), as well as cases and examples of analytics deployment in higher education. It should be clear at this stage that analytics in education have a diverse heritage and different disciplines are now converging around learning analytics. For example, it's not uncommon to see computer scientists, statisticians, psychologists, graphic designers, and visualization experts involved in an analytics project. Analytics is not yet a domain owned by a particular group of researchers. This is obviously an exciting prospect as there is greater inter-disciplinary discussion occurring in analytics than what is generally found in an established domain.

Learning analytics can be seen as a tiered or staged concept indicating progressive maturity of implementation.

  1. Extracting and analyzing data from learning management systems
  2. Building an analytics matrix that incorporates data from multiple sources (social media, LMS, student information systems, etc).
  3. Profile or model development of individual learners (across the analytics matrix)
  4. Predictive analytics: determining at-risk learner
  5. Automated intervention and adaptive analytics: i.e. the learner model should be updated rapidly to reflect near real-time learner success and activity so that decisions are not made on out-dated models
  6. Development of "intelligent curriculum" where learning content is semantically defined
  7. Personalization and adaptation of learning based on intelligent curriculum where content, activities, and socia connections can be presented to each learner based on her profile or existing knowledge
  8. Advanced assessment: comparing learner profile with architecture of knowledge in a domain for grading or assessment (see the image below taken from this article).

Within the 8 step model detailed above, dashboard, visualizations, and drill down reports are integrated at each level so that educators, learners, and administrators can explore and visualize the data.


This week, we will look at one specific aspect of the 8-step analytics model: semantically defined content. By defining the knowledge structure of a field of study (such as a key concept in introductory physics), we can begin to use automated evaluation of learner mastery of that content. By comparing what a learner knows and what constitutes competence in a domain of study, the learning process can be personalized and adapted.

Readings and Videos:

Semantic Web: An Introduction: http://infomesh.net/2001/swintro/

Ray, K (2009) Web 3.0 http://vimeo.com/11529540

Berners-Lee, T. (1989) Information Management: A proposal http://www.w3.org/History/1989/proposal.html

Tim Berners-Lee talk http://www.ted.com/talks/tim_berners_lee_on_the_next_web.html

Hilary Mason, Machine Learning: http://www.infoq.com/presentations/Machine-Learning

Jovanović, J., Gašević, D., Brooks, C., Devedžić, V., Hatala, M., Eap, T., Richards, G., "Using Semantic Web Technologies for the Analysis of Learning Content," IEEE Internet Computing, Vol. 11, No. 5, 2007, pp. 45-53, http://goo.gl/eouEW

Ali, L., Hatala, M. Gašević, D., Jovanović, J., "A Qualitative Evaluation of Evolution of a Learning Analytics Tool," Computers & Education, Vol. 58, No. 1, 2012, pp. 470-489, http://goo.gl/2gTd4

Activities this week:

On Tuesday (Valentine's day!) at 11 am Mountain time (conversions here:
http://www.timeanddate.com/worldclock/fixedtime.html?iso=20120214T11&p1=80), Dragan Gasevic will be leading a presentation on semantic data. The session will be held here: