Modles de Tlapprentsisage

1
SWEL04 Workshop Einhoven, August 23, 2004
Ontology-Based Referencing of Actors, Operation
and Resources in eLearning Systems
by Dr Gilbert Paquette CIRTA-LICEF Research
Center Télé-université, Montréal http//licef.telu
q.quebec.ca/gp
2
Backround Work at CIRTA-LICEF
AGD
1992-1995
1995-1997
MOT 2.0
AGD Advisor
MISA 2.0
3
Instructional Engineering

• A methodology that supports the analysis, the
  design and the delivery planning of a learning
  system, integrating the concepts, the processes
  and the principles of instructional design,
  software engineering and cognitive engineering

4
MISA 4.0 Method
5
MISA 4.0 Central Tasks
Instructional Modeling Axis
Knowledge Modeling Axis
212 Knowledge Model
214 Actors Competencies
222 Learning Events Network
310 Learning Units Content
410 Learning Instruments Content
320 Instructional Scenarios
6
MOT Graphic Language
7
Learning Events and Scenarios
8
IMS-Learning Design Information Model
9
Graphic Representation of a LD
10
A Simple IMS-LD Method
Course on
Ecological
Agriculture
C
C
Module B
Module A
(For agriculture
(For
expert)
beginners)
C
C
Act 1
Act 2
Synthetize
Analyze Gas
P
domains and
Properties
methods
C
C
C
Activity 4
Activity 5
Activity 1
C
C
Present a
Classify Gases by
Explore the
Systhesis on Gaz
their presence in
Agriculture
Properties
Agriculture
Domains and
Domain and
Activity 3
Methods
methods
Activity 2
Evaluate and
Make a
discuss
synthsis
synthesis
11
Improve Knowledge Referencing

• IMS-LD is a progress in eLearning specifications
• Some IMS-LD scenarios cannot be expressed in
  SCORM
• IMS-LD is more general, favors innovative methods
• Assigning optional objectives and prerequisites
  according to the IMS RDCEO specification (IMS
  2002) is not enough
• Consistency checking is not supported between
  levels nor between the content of learning
  activities and resources, and the actors
  competency
• Knowledge in learning resources is not described
• Actors knowledge and competencies is only
  indirectly defined through educational objectives
• Need for a qualitative structural representation
  of knowledge in activities, but also a
  quantitative one

12
Explor_at_-II Delivery System
13
Structured Competencies

• To say that somebody needs to acquire a certain
  knowledge is insufficient
• A skills taxonomy based on different viewpoints
  instructional objectives, generic
  tasks/processes, meta-knowledge
• Expandable taxonomy from general to specific
• Ordering skills from simple to complex
• Integrating domains of multiple intelligence
  cognitive, affective, social, psycho-motor

14
An integrated skills taxonomy
15
Skill/Performance Scale
Skills
Multimedia Production Method
Self-manage (10) Evaluate (9) Synthesize
(8) Repair (7) Analyze (6) Apply
(5) Transpose (4) Interpret (3) Identify
(2) Memorize (1) Pay attention (0)



.
Performance
Aware Familiarized
Productive Expert
16
Representing Knowledge in Explor_at_-II
17
A Multi Actor Workflow Interface
18
Functions Multi-Actor Processes

• Learning Design
• Learners assessment
• Knowledge processing
• Learning Object management
• Collaboration management
• Emerging activities management
• Delivery Models (Blended, Communities of
  Practice, EPSS,)
• ..

In LORNET, a TELOS LKMS is an aggregation of
FUNCTIONS
A function is a multi-actor process graph part
of a delivery model represeting a use-case of
the TELOS delivery system and a physiology of
the DLS
19
A Graphic OWL Editor
20
A Simple Ontology
21
Referencing with an Ontology
22
Skill/Performance Scale
Skills
Multimedia Production Method
Self-manage (10) Evaluate (9) Synthesize
(8) Repair (7) Analyze (6) Apply
(5) Transpose (4) Interpret (3) Identify
(2) Memorize (1) Pay attention (0)



.
Performance
Aware Familiarized
Productive Expert
23
Competency Equations
Components of a Function must reach competence
equilibrium . Ex Learning resources (persons,
documents and tools) must enable learners to
progress from an entry level to a target level
required by the activity.
24
Referencing Principles

• Tree organization of the knowledge referential
• allows competence inheritance from parent node to
  children
• reduce significantly the mechanisms of competence
  analysis and management.
• Must be completed by relational logic to sustain
  more refined mechanism of conceptual matching.
• Ontology referencing is insufficient without
  mastery levels
• prevent coarse granulation of sense
• weak semantic management services.
• Quantitative measures to weight ability on
  knowledge
• level scale to be reasonably simple, manageable
• levels correspondomg to clearly identify
  cognitive processes
• to be materialized as a learning design

25
LORNET Research Network
A major Canadian Initiative in Learning and
Semantic Web technologies

• 8 M in 5 years on learning and knowledge-based
  technologies
• Joint research with 27 industrial partners and
  organizations
• A team of 129 researchers, students and
  professionals.
• Consolidation of 7 research labs and centers and
  4 Research Chairs
• Convergence of complementary research areas
• Interaction with leading international projects
  OKI, IMS, CETIS, COLIS, Kaleidoscope

26
Future Research in LORNET

• New knowledge
• On Semantic Web techniques and applications
• On ICT support to Knowledge Management and
  training
• On learning objects creation, processing and use
• Tools and methods
• For ontologies creation and use
• For design, static and dynamic aggregation of LOs
• For content repurposing and adaptive assistance
• For knowledge extraction to model learning
  objects
• For advanced multimedia creation, search,
  delivery
• Systems
• TELOS Telelearning Operating System
• 5 LKMS and LKMA Learning and Knowledge
  Management Design and Delivery systems and
  application

27
SWEL04 Workshop Einhoven, August 23, 2004
Ontology-Based Referencing of Actors, Operation
and Resources in eLearning Systems
by Dr Gilbert Paquette CIRTA-LICEF Research
Center Télé-université, Montréal http//licef.telu
q.quebec.ca/gp