Transcript Combining ITS and eLearning – Opportunities and Challenges

Deutsches Forschungszentrum für Künstliche Intelligenz
Combining ITS and eLearning –
Opportunities and Challenges
Erica Melis, Jim Greer, Christopher Brooks, Carsten Ullrich
German Reseach Center for Artificial Intelligence (DFKI)
ARIES Lab, University of Saskatchewan, Canada
Why think about it again?
ITS
•
•
•
•
•
•
•
•
CogSci + AI community
Research-oriented
Lab evaluations
Restricted audience
“one-offs”
Few authors, designers
Individual tutoring
CSCL
eLearning
•
•
•
•
•
•
Web technology + commercial
Application-oriented
Very wide-scale usage
Broad audience
Standardization
Potential collaborative
authoring
• Efficient organization of
learning
• Usage of communication tools
Make eLearning more intelligent
Make ITS more open
ITS
+
+
+
+
+
+
+
+
-
Student modelling
Feedback generation
Diagnosis
Meta-cognitive support
Cognition and learning
Support of learning
activities
Special content
CSCL
Fix abstract domain map
Clearly defined audience
Single use
components/system
Little scalability
eLearning
+
+
+
+
+
+
+
OneSizeFitsAll
Simple feedback
Not much data
Little cognition and learning
Weak community tools
Massive multi-media
content
Content-based ontologies
Common communication
tools
Accessibility
Service-approach
Reusable components
Scalability
(Le)ActiveMath: eLearning
• Multi-lingual
• Integrates ITS components (Siette,
iCMap, ..)
• One central student model
• Integration with LMS
• Web presentation (of mathematics)
• KR: Semantic XML-representation
for maths+
• Semantic search
(Le)ActiveMath: ITS
•Adaptivity through instructional
planning
•Suggestions via „agents“
•KR: misconceptions,
competencies, dependencies
•Ontologies: content+topic map
• Individual feedback (authored and generated)
– Tutorial dialogues for symb differentiation
• Interactive concept mapping
• Lab evaluations
• Large classroom evaluations
ActiveMath: Contents
LeActiveMath Calculus
Universität Augsburg
University of Glasgow
150
students…
300 pages
de, en, es
Statistics
HTW Saarland
50 students,
200 pages
de
Optimization,
Operations Research
Mary State University
100 pages )
ru, en
3x50 students
1st year Calculus
U. Westminster, London
50 pages (exc)
en
250 students
Algebra Interactive!
Arjeh Cohen TUe
30 pages
en
Analysis Individuell
Uni Koblenz
Uni Saarland
20 students
300 pages
de
Matheführerschein
FH Dortmund
3 schools
50 pages
100 exercises
de
Fractions
Gesamtschule Bellevue
100 pages
70 students
de
iHelp: eLearning
•
•
•
•
•
•
•
Scorm compliant LMS
Powerful RBAC with IMS-SS
Wealth of learner data
Data mining
Semantic Web application
Integrates collaboration tools
Uses LMS as LOR
QuickT ime™ and a
T IFF (Uncompressed) decompressor
are needed to see t his picture.
Qui ckTime™ and a
TIFF (LZW) decompr essor
are needed to see this pi cture.
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Quic kTime™ a nd a
TIFF (Un co mp res sed ) d ec omp re sso r
ar e n eed ed to see thi s p ictu re.
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
QuickTime™ and a
TIFF(Uncompressed) decompressor
are needed to see thi s picture.
iHelp: ITS
• Data mining for metadata
• Ecological approach
– (learning objects in context)
• Selecting relevant learning objects
• Expertise location
• Adaptive content modules
• Awareness support
– for collaboration
– for competition
QuickT ime™ and a
T IFF (Uncompressed) decompressor
are needed to see t his picture.
Qui ckTime™ and a
TIFF (LZW) decompr essor
are needed to see this pi cture.
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Quic kTime™ a nd a
TIFF (Un co mp res sed ) d ec omp re sso r
ar e n eed ed to see thi s p ictu re.
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
QuickTime™ and a
TIFF(Uncompressed) decompressor
are needed to see thi s picture.
Challenges: reuse and interoperability
•
•
•
•
•
Distributed architecture
Distributed content
Collaborative authoring
Make available and reuse domain reasoners
Use blackbox services
Challenges: knowledge representation
•
•
•
•
•
Collaborative authoring…
2-level ontologies
Ontology-mappings
Ontologies in context, contextualized metadata
Semantic versioning
• Learning metadata
• Validate metadata
• Extensibility of representations
Challenges
ITS
Decision Analysis
Pedagogical
Principles
Semantic Web
Intelligent
LMS
Recommender Systems
Learning Object
& Metadata
Instructional Design
Challenges: ??
• Support critical thinking
• And other meta-cognitive
competencies
• Self-organzational models
of learning
Conclusion
• Mutual benefits and chances to get widely used
• New challenges
• Bridge the gap!
OMDoc Knowledge Representation
<definition id="monoid/def_monoid" for="monoid"
<metadata>
<depends-on>
<xref theory="structures/structure" />
</depends-on>
<Title xml:lang="en">Definition of a monoid</Title>
</metadata>
<CMP xml:lang="en" format="omtext">
A monoid is a <ref xref="structures/def_structure"> structure </ref>
<OMOBJ> <OMA>
<OMS cd="elementary" name="ordered-triple"/>
<OMV name="M"/> <OMS cd="semigroups" name="times"/> <OMS cd="semigroups" name="unit"/>
</OMA></OMOBJ>
in which
<OMOBJ> <OMA>
<OMS cd="elementary" name="ordered-pair"/>
<OMV name="M"/> <OMS cd="semigroups" name="times"/>
</OMA> </OMOBJ>
is a semi-group
with <ref xref="semigroups/def_unit">e</ref>
<OMOBJ>
<OMS cd="semigroups" name="unit"/>
</OMOBJ>.
</CMP>
<FMP><OMOBJ> ... </OMOBJ></FMP>
</definition>