Plan

TICL-08 Symposium, New-York, March 25 2007

Ontology Modeling for Comptency-Based Learning

Environments

Gilbert Paquette

Director of the CICE Canada Research Chair LICEF Research Center, Télé-université

www.licef.teluq.uquebec.ca/gp

Plan

1. Backround: Knowledge-based ID and Semi-formal modeling

2. Competencies for Structuring KB Learning Environments

3. Competencies as Meta-processes and Strategies for Learning Scenarios

4. Ontology for Referencing Resources

5. Activity Assistance based on Meta-process Principles and Domain ontology

6. Knowledge Representation Principles

1- Background

MOT +

MOT 2.0

1995-19971995-1997

AGD

1992-19951992-1995

MISA 3.0

MISA 2.0

1995-19971995-1997

MISA 4.0

MISA LD

MISA forms

1997-19981997-1998

ADISA/Explor@

1999-20021999-2002

TELOSScenario Ed.Ontology Ed.

2006-20072006-2007

1998-19991998-1999

MOT+LD

2004-20052004-2005

MOT+OWL2005-20062005-2006

ID MethologyModeling Tools

eLearningSystems

Use in Instructional Engineering (MISA)

640 Maintenance/Quality Management

630 Learning System/Resource Management

620 Actors and Group Management

610 Knowledge/ Competency Management

Phase 6 – Phase 6 – Delivery PlanDelivery Plan

540 Test Planning 542 Revision Decision LogPhase 5 – Val.Phase 5 – Val.

440 Delivery Models

442 Actors and their resources

444 Tools and Telecom446 Delivery Services

and Locations

430 Learning Resource List

432 Media Models 434 Media Elements 436 Source Documents

420 Learning Resource Properties

410 Learning Resource Content

Phase 4 –Phase 4 –Detailed Detailed DesignDesign

340 Delivery Planning330 Development Infrastructure

320 Learning Scenarios

322 Activity Properties

310 Learning Unit Content

Phase 3 –Phase 3 –ArchitectureArchitecture

240 Delivery Principles 242 Cost-Benefit

Analysis

230 Media Principles220 Instructional Principles

222 Event Network224 Learning Unit

Properties

210 Knowledge Model Orientation Principles

212 Knowledge Model

214 Competencies

Phase 2 – Phase 2 – Initial solutionInitial solution

Delivery AxisDelivery AxisMedia AxisMedia AxisPedagogy Pedagogy AxisAxis

Knowledge Knowledge AxisAxis

100 Organization’s Training System 102 Training Objectives 104 Learners’ properties106 Present Situation 108 Reference Documents

Phase 1- Phase 1- DefinitionDefinition

MOT Semi-formal Modeling

Taxonomy of knowlege models

KnowledgeModels

Factual Models

Set of Examples

Set of Statements

Conceptual Models

Typologies

Component Systems

Hybrid Conceptual

Systems

Procedural Models

Series Procedures

Parallel Procedures

Iterative Procedures

Prescriptive Models

Norms and Constraints

Laws and Theories

Decision Trees

Control Rules

Processesand Methods

Processes

Methods

Multi-actorworkflows

Set of traces

Goals for a RepresentationLanguage

Transparent semantic to facilitate design and Transparent semantic to facilitate design and communication at an informal levelcommunication at an informal level

Integrated representation for Concept Maps, Flow Integrated representation for Concept Maps, Flow Charts, Decision Trees and others. Charts, Decision Trees and others.

Generality : domains, types of models, granularity, Generality : domains, types of models, granularity, higher level knowledgehigher level knowledge

From Semi-formal to Formal RepresentationFrom Semi-formal to Formal Representation

Informal Semi-formal FormalWritten-Oral

CommunicationUML Diagram

MOT KnowledgeModels

Conceptual graphsOntologies (MOT+OWL)Rules and Constraints

MOT+OWL: A Formal Graphic Ontology Editor

RelationalProperty

owl:Class3> <owl:intersectionOf rdf:parseType="Collection">List of class descriptions </owl:intersectionOf></owl:Class3>

Class intersection x: Class3(x) Class1(x) Class2(x)

2- Knowledge Management:Enhancing Human Competency

Goal: knowledge and competency sharingCompetency implies higher level knowledge apply to domain knowledgeStructured competencies: knowledge, skills/attitude and performance/context of use.

COMPETENCY

1. Knowledge 2. Generic Skill3. Performance Context

C CC

Select in a domain ontology

I/P

Select in a Skill’s taxonomy

CombinePerformance/ context

criteria

I/P I/PScale position

C C

Combining viewpoints : instructional objectives (Bloom) generic tasks (Chandrasekaran) meta-knowledge (Pitrat)

Generic Skills Taxonomy

Identify

Illustrate

Memorize

Utilize Classify

Construct

Initiate/ Influence

Adapt/ control

Discriminate

Explicitate

SimulateDeduce

Predict

Diagnose

Induce

Plan

S

Exerce a skill

Receive

Reproduce

S

Create

Self- manage

S

S

1-Show awareness

S

9-Evaluate

S

4-Transpose

S

7-Repair

S

2-Internalize

S

3-Instantiate /Detail

S

5-Apply

S

6-Analyze

8-SynthesizeS

S

10-Self- manage

S

Generic skill Inputs Products

Simulate Process to simulate: inputs, products, sub-procedures, control principles

Trace of the procedure: set of facts obtained through the application of the procedure in a particular case

Construct Definition constraints to be satisfied such as target inputs, products or steps….

A model of the process: its inputs, products, sub-procedures each with their own inputs, products and control principles

3- Generic Simulation Strategy

(5) Simulation

meta-process

Produce examples of the input concepts

Identify the next applicable

procedure

Execute the procedure using its

execution principles

Assemble the simulation

trace

Description of the process to be

simulated

Inputs to the simulated

process

Products of the procedure

Simulation trace of

the procedure

Execution principles of

the simulated procedure

More procedures to execute

No more procedures to

execute

I/P I/P

I/P

I/P

I/P

P

P P

P

I/P

I/P

C

C C

C

p

I/P

I/P

DescriptionPrinciples

PresentationPrinciples

ExamplegenerationPrinciples

Procedureidentification

Principles

CompletenessPrinciples

R

RR

R

R

I/P

C

Assisted Simulation Scenario(Multimedia Production Domain)

DesignerCase studies for a method to select

procedures

Interactions on examples processed by

learners

Text presenting examples of simulations

I/P

I/P

I/P

Prepare learning materials

RI/P

I/PI/P

Content expert

Learner/expert

Interactions

I/P

Interact by email

R

I/P

R

Trainer

Presentation and discussion of completeness

principles

FAQ onpresentation

norms

I/P

I/P

Use a forum software

Maintain a FAQI/P

R I/P

Activity 1: Choose a MM

process tosimulate

Activity 2: Choose a typical

multimedia project

Activity 3: Identify a MM

production task

Activity 4: Execute a

production task

Activity 5: Verify is the

process is complete

Activity 6: Produce a project report on

the MM process

Assistance agent

Learner/Agent

Interactions

I/P

Interact in scenario

R

I/P

4- Referencing Resources with Ontologies

5- Assistance Methodology1.1. Define target competencies: generic Define target competencies: generic

process and domain knowledge ontologyprocess and domain knowledge ontology2.2. Define executable scenario (task structure of Define executable scenario (task structure of

the host environment) the host environment) 3.3. Add assistance objects to critical tasksAdd assistance objects to critical tasks4.4. Integrate assistance: for each critical task Integrate assistance: for each critical task

define product attributes and progression define product attributes and progression levelslevels

5.5. Define conditions and actions based on the Define conditions and actions based on the relation between input knowledge and relation between input knowledge and product attributeproduct attribute

Assistancetree

Tasktree

Input-Outputs

6. Properties of the Knowledge Representation Paradigm

GraphicGraphic. Reduce ambiguity by the use of . Reduce ambiguity by the use of standardized objects and links.standardized objects and links.User-friendliness.User-friendliness. Typed links are preferred, Typed links are preferred, not two few nor two many types of links, clear not two few nor two many types of links, clear semantic.semantic.General.General. Capacity to represent knowledge in Capacity to represent knowledge in very different subject domains, at various levels very different subject domains, at various levels of granularity and precisionof granularity and precisionFormalizableFormalizable. Upward compatible from informal . Upward compatible from informal graphs, up to semi-formal and totally graphs, up to semi-formal and totally unambiguous formal models.unambiguous formal models.

Properties of the Knowledge Representation Paradigm (cont’d)

DeclarativeDeclarative. Separates knowledge from their . Separates knowledge from their processing. Describe processing knowledge processing. Describe processing knowledge declaratively, so that higher order meta-declaratively, so that higher order meta-knowledge, applies to specific knowledge.knowledge, applies to specific knowledge.

StandardizedStandardized. To enlarge communication . To enlarge communication between persons and/or software agents.between persons and/or software agents.

Computable.Computable. Formal representation that can be Formal representation that can be processed by computer agents, in a complete processed by computer agents, in a complete and decidable way (e.g. OWL-DL).and decidable way (e.g. OWL-DL).

TICL-08 Symposium, New-York, March 25 2007

Ontology Modeling for Comptency-Based Learning

Environments

Gilbert Paquette

Director of the CICE Canada Research Chair LICEF Research Center, Télé-université

www.licef.teluq.uquebec.ca/gp