Title: Ontologies for Effective Use of Context in e-Learning Settings
1Ontologies for Effective Use of Context in
e-Learning Settings
- Dragan Gaševic1, Jelena Jovanovic2,
- Colin Knight3, Griff Richards3
- 1School of Computing and Information System,
Athabasca University - dgasevic_at_acm.org
- 2FON-School of Business Administration,
University of Belgrade - jeljov_at_gmail.com
- 3School of Interactive Arts and Technology, Simon
Fraser University - cjk, griff_at_sfu.ca
2What we do not have today and we need
Individualized personalized learning supported
by ICT Adaptive and contextualized learning
experience supported by ICT Effective and
efficient information, community, and expert
retrieval, based on context and
semantics Conceptual modeling of different
knowledge domains
(Roadmap for Technology Enhanced Professional
Learning, 2007)
3What we do not have today and we need
Individualized personalized learning supported
by ICT Adaptive and contextualized learning
experience supported by ICT Effective and
efficient information, community, and expert
retrieval, based on context and
semantics Conceptual modeling of different
knowledge domains
(Roadmap for Technology Enhanced Professional
Learning, 2007)
4Existing Solutions
- Learning metadata specifications and standards
- IEEE LOM standard
- IMS Learning Design specification
- Problems with existing solutions
- Prescriptive in nature
- Metadata of limited quantity, complexity and
semantics - No room for metadata validation and revision
- Ambiguous provenance of metadata
- One-way (meta)data flow
5Existing Solutions
- Alternative solutions offered by the research
community - Ecological approach Pragmatic Web (G.
McCalla, USASK, Canada) - Use of concepts from controlled vocabularies,
taxonomies, and ontologies as metadata values - How to get closer to the users?
6Our Solution
- Changing the concept and the target users of
metadata - metadata derived from the actual usage of LOs
- metadata aimed at computer programs that can make
sense of them for end users - An ontology based framework for capturing and
representation of contextualized semantic rich
metadata
7Learning Object Context
- Context awareness (meta)data
- Learning activity a learning object (LO) was used
in - Prerequisites, learning objectives, available
time, - Domain topics/concepts covered,
- LOs components of special interest,
- Pedagogical role played by the LO,
- Learners evaluations of the LO,
- Learners characteristics
- Aimed at capturing diverse kinds of learning
situations typical for modern e-Learning
environments
Learning Object Context
8Learning Object Context Ontologies (LOCO)
Framework
- Formalizes the notion of Learning Object Context
- Integrates different kinds of learning related
ontologies - Flexible and easily extensible
9The LOCO Framework
- Aimed at facilitating
- Reuse/repurposing of LOs and learning design
- Context-aware retrieval of learning resources
- Context aware recommendation of learning
content/activities/peer students
Won us The Best Paper Award at the 6th IEEE
Intl Conference on Advanced Learning
Technologies (ICALT2006)
10Learner-centric scenarios
Providing learner with a custom view of the LOs
repository, compliant with the requirements of
the current learning context
11Learner-centric scenarios
Leveraging context (meta)data of learners
on-line communication and collaboration
12Teacher-centric scenarios
- Provision of fine grained and semantically rich
feedback for educators, e.g., information about - domain topics difficult for learners
- unusual performance of a learner or a group of
learners - The idea is to help educators
- rethink the quality of
- the learning content and
- learning design of the online courses they teach
- make more informed decisions regarding how to
improve their online courses
13An Application Example TANGRAM
- Ancient Chinese moving piece puzzle
- An adaptive learning environment for the domain
of Intelligent Information Systems - Two basic functionalities
- On-the-fly assembly of personalized learning
content out of existing content units - Quick access to a particular type of content
about a topic of interest - e.g. access to examples of RDF documents or
definitions of the Semantic Web.
14An Application Example TANGRAM
- TANGRAM leverages the LOCO framework
- Decomposes LOs in accordance with a content
structure ontology - enables direct access to and reuse of content
units of diverse granularity levels - Annotates content units with concepts of an
ontology of instructional roles - Annotates content units with concepts of the
domain ontology - Keeps track of its users through the concepts and
properties of its user model ontology
15An Application Example TANGRAM
16An Application Example TANGRAM
- Weaknesses to be addressed
- Relies on a rudimentary form of learning design
ontology - Lacks support for active and collaborative
learning
To learn more about TANGRAM visit
http//iis.fon.bg.ac.yu/TANGRAM/home.html
17An Application Example LOCO-Analyst
- Aims at helping educators rethink the content and
design of their courses - Provides educators with feedback about
- the learning activities of their students
- the usage of the deployed learning content
- the peculiarities of the interactions in the
online learning community. - Not tied to any specific e-Learning environment
- feedback provision functionalities developed on
top of the LOCO framework
18An Application Example LOCO-Analyst
- Based on Semantic Web technologies
- Ontologies enable
- unambiguous representation and
- integration of learning context data
- abstraction of relevant concepts from the usage
tracking data - Semantic annotation facilitates establishing
semantic connections among learning artifacts - Simple rules are used for feedback generation
19An Application Example LOCO-Analyst
20An Application Example LOCO-Analyst
- Evaluated as a useful tool for improving content
and design of online courses - Special appreciation for
- Qualitative (over quantitative) feedback
- Integrated view on the learning process as a
whole - Expressed an interest in using LOCO-Analyst in
ongoing course delivery
To learn more about LOCO-Analyst visit
http//iis.fon.bg.ac.yu/LOCO-Analyst/
Please, ask for a demo today ?.
21LOCO-Analyst Current Work
Reducing the overhead in domain ontology
creation/evolution by leveraging students
collaborative tagging
22Conclusion
- Peer-review
- Courses
- Publishing
- Portfolio
- Evidence
- Competencies
Students/Educators
- Community
- Course, University,
- Content
- LORs
- Libraries
- Multimedia
- Reuse
- Collaboration
- Chat
- Discussion
- Services
- Authoring
- Word
- Frontpage
- Reload
- Pedagogy
- User models
- Adaptivity
- Educational models
23Conclusion
User modeling
Content Structure
ContentType
Domain
- Peer-review
- Courses
- Publishing
- Portfolio
- Evidence
- Competencies
- Domain tools
- Mashed up with education tools
Students/Educators
- Community
- Course, University,
- Content
- LORs
- Libraries
- Multimedia
- Reuse
- Collaboration
- Chat
- Discussion
- Services
- Authoring
- Word
- Frontpage
- Reload
- Pedagogy
- User models
- Adaptivity
- Educational models
24Conclusion
- LOCO-Analyst Achieved results
User Model
LearningContext
Content Structure
ContentType
Domain
- Peer-review
- Courses
- Publishing
- Portfolio
- Evidence
- Competencies
- Domain tools
- Mashed up with education tools
Students/Educators
- Community
- Course, University,
- Content
- LORs
- Libraries
- Multimedia
- Reuse
- Collaboration
- Chat
- Discussion
- Services
- Authoring
- Word
- Frontpage
- Reload
- Pedagogy
- User models
- Adaptivity
- Educational models
25Future Ongoing Work Learning Knowledge Landscape
26Learning Knowledge Landscape
User-centered
27Thank you!
28The LOCO-Cite ontology
29The LOCO-Cite ontology
30The LOCO-Cite ontology
31The LOCO-Cite ontology