1212 department of computer science writing mot, reading aha! - converting between an authoring and...

Department of Computer Science

Writing MOT, Reading AHA! - converting between an authoring and a delivery system for adaptive educational

hypermedia -

Alexandra Cristea, David Smits and Paul de Bra

[email protected], [email protected]


Outline1. Motivation2. MOT authoring (according to LAOS):

1. DM2. GM3. UM4. PM5. AM

3. CAF: The Common Format4. Delivering in AHA!5. Conclusion


Motivation

• authoring static educational content is quite different from adaptive content

• growing understanding: a great deal of help is necessary for authors

• to make authoring easier (‘authoring once, delivering many’): 2 major approaches: – a common language (lingua franca) used by

all authors of AEH– the use of converters between AEHs


DM(DomainModel)


GM (goal and constraints model)


UM (user model)

• typical variables: – knowledge level, – interests, – learning styles


UM – cont.

• In MOT, the LAG language is used:• Independent user map variables declaration:

// VARS UM.GM.stereotype1• Overlay user map variables declaration:

// VARS UM.GM.Concept.knowledge• Independent user map variables initialization:

initialization ( UM.GM.stereotype1 = beg )

• Overlay user map variables initialization: initialization ( // … other initializationswhile UM.GM.Concept.knowledge != 0

( UM.GM.Concept.knowledge = 0)) // end initialization


PM (presentation/ physical model)

• provide bridge to actual code generation for different platforms

• physical properties and environment of presentation: – ideal page length– where presentation chapters should be cut into pages– how and where multimedia should appear (p.o.v. display

possibilities) – colours– fonts, etc. – but also: server load, traffic decisions, compression, etc.

• in a simplified authoring experience: – generated automatically, to make the authoring easier


PM – cont.

• In MOT, the LAG language is used:• Independent PM variables declaration:

// VARS PM.GM.Menu, PM.GM.ToDo, PM.GM.Next

• Overlay PM variables declaration: // VARS PM.GM.Concept.show

• Independent presentation map variables initialization: initialization ( // … other initializationsPM.GM.Menu = falsePM.GM.ToDo = false // … other initializations)

• Overlay presentation map variables initialization: initialization ( // … other initializationswhile GM.Concept.label = null ( PM.GM.Concept.show = true ) // … other initializations)


AM

• Uses LAG – a classification method for adaptive

techniques as • direct adaptation rules, • adaptation language and • adaptation strategies.

– repartition on layers follows increasing level of semantics.

• MOT is implementing (a v. of) all three levels


AM: assembly level rules

• generic:if GM.Concept.weight > 10 then ( PM.GM.Concept.show = true )

• specific:if '\Main topic\video presentation.title'.access

== truethen ( UM.GM.stereotype1 = vis )


AM: adaptation language level rules • repetitive actions:

while GM.Concept.label == null ( PM.GM.Concept.show = true )

• enough conditions to proceed to next level: if enough( GM.Concept.label == int UM.GM.stereotype1 == int , 2 ) then ( PM.GM.Concept.show = true )


AM: adaptation language level rules – cont.

• structural semantics from the domain or lesson maps:

• Type of Attributes (in domain maps):DM.Concept.Attribute.Type == Title

• Order of Concepts (in lesson groups):GM.Concept.Order == 1

• Level of a Concept (in lesson hierarchy):GM.Concept.level == 2

• Parent/ Child of a concept (in lesson hierarchy):GM.Concept.parent / child


AM: adaptation strategies

• LAG program: • // Description• // Variables• initialization ()• implementation ()


CAF:<?xml version="1.0" encoding="UTF-8"?><!ELEMENT CAF (domainmodel?, goalmodel?)><!ELEMENT domainmodel (concept+)><!ELEMENT concept (name, attribute*, concept*)><!ELEMENT attribute (name, contents)><!ELEMENT name (#PCDATA)><!ELEMENT contents (#PCDATA)><!ATTLIST contents weight CDATA "" label CDATA ""><!ELEMENT goalmodel (lesson)><!ELEMENT lesson (contents*, lesson*)>


CAF: DM<CAF><domainmodel> <concept> <name>Adaptive</name> <concept>

<name>Adaptive HyperMedia</name><attribute> <name>title</name> <contents>Adaptive HyperMedia</contents></attribute>

…</concept>

…</domainmodel></CAF>


CAF:GM<CAF> ...<goalmodel> <lesson> <contents weight="0" label="">Adaptive\Adaptive HyperMedia\title</contents> <contents weight="0" label="">Adaptive\Adaptive HyperMedia\text</contents> <lesson>

<contents weight="0" label="">Adaptive\Adaptive HyperMedia\Welcome\title </contents>

<contents weight="0" label="vid">Adaptive\Adaptive HyperMedia\Welcome\video

</contents> </lesson> ...</lesson><goalmodel></CAF>


Delivering in AHA!


Conclusions

• one of the 1st attempts to convert info between 2 different AEHS: – an authoring system & a delivery system

• evaluated within a class of 4th year undergraduate students of CS at Eindhoven University of Technology

• yet another step to establish a common platform and a common format for the representation of AEH– From our first tests with different systems it is clear

that this common format has to have a rich semantics


Try it out at:

• http://wwwis.win.tue.nl/~acristea/mot.html

1212 department of computer science writing mot, reading aha! - converting between an authoring and...

Documents