how recommender systems in technology-enhanced learning depend on context
DESCRIPTION
Presentation at STELLAR - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DETRANSCRIPT
How Recommender Systems in
Technology-Enhanced Learning
depend on ContextHendrik Drachsler Open University of the Netherlands
Nikos ManouselisGreek ResearchTechnology Network
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 2 | November 30, 2009
TEL context
Figure by : Cross, J. (2006). Informal learning: Rediscovering the natural pathways that inspire innovation and performance. San Francisco, CA: Pfeiffer.
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 3 | November 30, 2009
Formal learning
Figure by:Cross, J. (2006)
• are learning offers from educational institutions.
• is imbedded into a curriculum or syllabus framework.
• is highly structured.• leads to a specific accreditation.• involves domain experts to guarantee quality.
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 4 | November 30, 2009
Formal learning = structured layers
Generic layers within a simplified architecture of an educational AEH(Karampiperis & Sampson, 2005)
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 5 | November 30, 2009
Informal learning• content is provide from different sources.
• happens outside formal educational settings (e.g. related to work or leisure time.
• is less structured (in terms of learning goals, study time or learning support).
• does not lead to a certain accreditation.
Figure by:Cross, J. (2006)
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 6 | November 30, 2009
Informal learning = emergence
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 7 | November 30, 2009
Context variables Formal learning Curriculum (Closed-Corpus) Teacher directed Predefined learning resources, learning goals Maintenance
Informal learning Learning resources from different providers (Open-Corpus)
More self-directed learning goals
Responsible for own learning pace / path
Lack of maintenance
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 8 | November 30, 2009
Recommendation approaches
Learning settings, environmental conditions and the task greatly affect the design of systems in TEL.
Learning settings, environmental conditions and the task greatly affect the design of recommender systems in TEL.
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 9 | November 30, 2009
Formal recommendation approach
Content Layer
Conceptual Layer
Learning Goal LayerAdaptiveSequencing(top-down)
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 10 | November 30, 2009
Informal recommendation approach
Content Layer
Clustering Layer 1..n
Clustering Layer nHierarchicalclustering(bottom-up)
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 11 | November 30, 2009
Research question
How can we get the best out of both worlds?
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 12 | November 30, 2009
A solution for formal learning
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 13 | November 30, 2009
• 100 user (hopefully some more after today )
• 20000 Web 2.0 items (increasing every hour)
• 700 ratings in the data base
• 10000 tags in the data base
A solution for informal learning
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 14 | November 30, 2009
Version 1.0
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 15 | November 30, 2009
Version 1.0
DUINE Prediction Engine
Database of Items
User Interface
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 16 | November 30, 2009
How does it work?
Cold-Start = Tag-based recommendation
Collaborative Filtering with ratings
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 17 | November 30, 2009
Learner profile
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 18 | November 30, 2009
Conclusions
Fed by bottom-upapproach
Fed by top-downapproach
How tocombine?
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 19 | November 30, 2009
Future R&D
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 20 | November 30, 2009
Many thanks for your interest!
This slide is available here:http://www.slideshare.com/Drachsler
Email: [email protected]: celstec-hendrik.drachslerBlogging at: http://elgg.ou.nl/hdr/weblogTwittering at: http://twitter.com/HDrachsler
[email protected] - Alpine rendez-vous workshop on context-aware recommendation 2009, Garmisch-Partenkirchen, DEPage 21 | November 30, 2009
You can use it as well!
Register atremashed.ou.nl.
Enter your favorite Web 2.0 potatoes.
Join the community.
ReMashed starts mashing.
Taste your personal flavor of Web 2.0.
http://remashed.ou.nl
Please sign up at: