AN AUTHORING ARCHITECTURE FOR ANNOTATING

EDUCATIONAL CONTENTS

Domain, Sequencing and Content-Repository Ontologies

José M. Gascueña, Antonio Fernández-Caballero and Pascual Gónzalez

Universidad de Castilla-La Mancha, Departamento de Sistemas Informáticos &

Instituto de Investigación en Informática de Albacete, Campus Universitario s/n, 02071-Albacete, Spain

Keywords: E-learning, Ontology, Learning objects.

Abstract: E-learning platforms available nowadays are mainly centred in supporting management tasks, but they do

not include or even consider in a too satisfactory way the adaptation to student’s profile, the reusability of

educational materials, or the efficient search into educational materials. By combining the paradigms of

ontologies and learning objects in authoring tools it is possible to annotate educational contents for

generating personalized material. The characteristics introduced in this paper are the learning style best

suited to the student, the device used to access the contents and the skill to be developed when using the

material. The general architecture of the proposed tool is fundamentally composed of three different and

interrelated ontologies: domain, sequencing and content-repository ontologies, where all knowledge about

which educative content is taught, how it is taught and how it is organized is respectively stored.

1 INTRODUCTION

Nowadays a great number of e-learning platforms

are available (e.g. WebCT, Moodle, ATutor). They

are mainly centred in supporting management tasks

such as to administrate users and groups, or to store

educational materials (Shimic, Gasevic & Devedzic,

2006). Nevertheless, these platforms generally do

not include or even consider in a too satisfactory

way some important issues. For instance, they offer

the same materials and activities to all students; thus,

the content shown is not always adapted to their

knowledge, preferences and objectives, that is, to the

students profiles. Moreover, there are few

possibilities of reusing the educational materials due

to their low granularity. Indeed, frequently it is not

possible to make a search directly into files

fragments (e.g. an image, a table, a graph, a schema,

or a summary) without having to open and review

the documents offered.

In this paper, an ontology-based authoring

architecture for annotating educational contents is

proposed to solve some of the problems commented

above. The architecture generates adaptable material

according to characteristics such as the most

appropriate learning style for a given student, the

device used to access to contents, and the skill that is

sought to be developed when using it.

An ontology is an explicit specification of a

conceptualization (Gruber, 1993); it allows defining

explicitly and formally the concepts and relations

that appear in the application domain. The use of

ontologies in the e-learning community is useful for

several reasons. (1) It helps authors in building

consistent and well-elaborated material, (2) it

provides facilities to construct enhanced search

engines for finding more relevant learning material,

(3) it enables individual adaptive delivery of

learning services as well as dynamic navigational

support, and, (4) it provides the means for

constructing a reference model for learning

resources (Leidig, 2001).

In our research only resources that may be

transmitted through the Internet are considered.

Thus, among the different definitions found of the

concept of learning object (McGreal, 2004), the

more appropriate definition to our objectives is the

one provided by Wiley (2001): “Any digital resource

that can be reused to support learning”. The

reusability characteristic inherent to learning objects

is one of the reasons for the increasing interest in

adopting this paradigm. It decreases time and effort

required to produce learning objects in a significant

M. Gascueña J., Fernández-Caballero A. and Gónzalez P. (2007).

AN AUTHORING ARCHITECTURE FOR ANNOTATING EDUCATIONAL CONTENTS - Domain, Sequencing and Content-Repository Ontologies.

In Proceedings of the Ninth International Conference on Enterprise Information Systems - HCI, pages 53-60

DOI: 10.5220/0002353900530060

 SciTePress

manner. We consider all the resources defined at the

level of paragraph, image, table, diagram, audio, and

so on, as learning objects to obtain a high grade of

reusability.

In our architecture, learning objects are labelled

according to learning styles theory in order to allow

delivering didactic materials suitable to all students’

learning styles. Due to the current proliferation of

new and different devices, it seemed also useful to

present different content formats, based on

hardware, software and network connection features

of the devices used by the students.

The World Wide Web Consortium (W3C) has

developed OWL (Web Ontology Language) (Dean

et al., 2004) to increase an expressive power that the

earlier existing ontology markup languages – XML,

RDF and RDF-Schema, among others – did not

have. For instance, OWL incorporates Boolean

combinations of classes (union, intersection, and

complement), disjointness of classes, cardinality

restrictions, special characteristics of properties

(transitive, unique, inverse), and local scope of the

properties (Antoniou & van Harmelen, 2004). All

OWL ontologies introduced in this paper are

encoded in OWL language and the Protégé

framework has been selected to edit/construct them.

The article is structured as indicated next. In

section 2 the works carried out in the last years

related to the learning objects paradigm and based

on ontologies are described. In section 3 the

proposed architecture for annotating educational

contents is introduced in extensive. Finally, some

conclusions are offered.

2 RELATED WORK

In scientific literature several proposals related to the

use of ontologies in educational environments may

easily be found. An ontology-based metadata to

achieve personalization and reuse of content in

AdaptWeb project has been described (Silva &

Palazzo, 2004). In mentioned project, DAML+OIL

language is used to represent the ontology; but

unfortunately, an inexpert user in ontologies is not

provided with any user friendly interface tool to

populate it. The architecture of the adaptive learning

system proposed in Duitama (2005) is based on the

AHAM reference model for adaptive hypermedia

applications, and RDF Schema is used to represent

knowledge models that compose it, which again

entails the previously commented limitations.

Another approach (Santacruz-Valencia et al., 2005)

provides ontology-based mechanisms to carry out

the leaning objects assembly, where the knowledge

(requirements and competencies) associated to

learning objects is kept in mind to determine if it is

possible to assemble them. In this work features

such as learning style, software and hardware are not

considered for the purpose of the assembly.

TANGRAM is a learning environment in the domain

of Intelligent Information Systems (IIS) useful to

authors and students interested in this domain. In

this case, Semantic Web technologies and ontologies

in particular are used. This environment is centred in

a concrete domain, so it is not enough generic

(Jovanovic, Gasevic & Devedzic, 2006). In another

work (Baloian et al., 2004) a mechanism for

information retrieval not only taking into account the

student profiles but the equipment issues is

presented. Lastly, in (Ronchetti & Saini, 2004) an

architecture to help students to find materials that

present different points of view or different ways to

explain concepts is proposed, but again it does not

make use of Semantic Web technologies.

3 FUNCTIONAL DESCRIPTION

The proposed architecture is generic, that is to say, it

is not associated to a course in particular. But rather

it defines the characteristics that appear in any

course. The aim is to create an “instance” for each

particular course from the general ontologies

(mainly, sequencing ontology and domain ontology,

as described next) included in the tool.

Basically, the authoring tool architecture

proposed consists of several distinct ontologies (see

Figure 1). The domain ontology describes concepts,

and relations between them, that appear in the

domain of a course. The sequencing ontology

defines the possible learning paths that can be given

through the concepts defined in the domain

ontology. The contents-repository ontology includes

metadata to describe learning objects, as well as the

relationships between the different kinds of objects,

used to teach the concepts belonging to the courses

inserted with the tool. In this ontology, besides the

definition of classes and relations among them, there

will be individuals when the ontology is populated

with the authoring tool. The mapping ontology

allows establishing relations between the concepts

(instances) included in the domain ontologies of

different courses. Lastly, the administration

ontology includes information about the authors of

the contents (personal information and courses that

they are authorized to access) and where the

ontologies included in the tool are located.

ICEIS 2007 - International Conference on Enterprise Information Systems

S e q u e n c i n g O n t o l o g y

D o m a i n O n t o l o g y

C o n t e n t s - R e p o s i t o r y O n t o l o g y

Course 1

Sequencing

Ontology

Course 2

Sequencing

Ontology

Course N

Sequencing

Ontology

Course 1

Domain

Ontology

Course 2

Domain

Ontology

Course N

Domain

Ontology

C1-C2

Mapping

Ontology

C1-CN

Mapping

Ontology

importimportimport

import

importimport

C2-CN

Mapping

Ontology

import import import

import

importimport

import

A u t h o r i n g E n v i r o n m e n t

Sequencing

Tool

Domain

Tool

Repository Management

Tool

U s e r

Contents

A d m i n i s t r a t i o n O n t o l o g y

(a u t h o r s a n d o n t o l o g i e s m a n a g e m e n t)

A d m i n i s t r a t i o n O n t o l o g y

(a u t h o r s a n d o n t o l o g i e s m a n a g e m e n t)

S e q u e n c i n g O n t o l o g y

D o m a i n O n t o l o g y

C o n t e n t s - R e p o s i t o r y O n t o l o g y

Course 1

Sequencing

Ontology

Course 2

Sequencing

Ontology

Course N

Sequencing

Ontology

Course 1

Domain

Ontology

Course 2

Domain

Ontology

Course N

Domain

Ontology

C1-C2

Mapping

Ontology

C1-CN

Mapping

Ontology

importimportimport

import

importimport

C2-CN

Mapping

Ontology

import import import

import

importimport

import

A u t h o r i n g E n v i r o n m e n t

Sequencing

Tool

Domain

Tool

Repository Management

Tool

A u t h o r i n g E n v i r o n m e n t

Sequencing

Tool

Domain

Tool

Repository Management

Tool

U s e r

ContentsContents

A d m i n i s t r a t i o n O n t o l o g y

(a u t h o r s a n d o n t o l o g i e s m a n a g e m e n t)

A d m i n i s t r a t i o n O n t o l o g y

(a u t h o r s a n d o n t o l o g i e s m a n a g e m e n t)

Figure 1: General layout of the authoring tool.

The courses managed with the tool are produced

by creating particular sequencing and domain

ontologies for each included course, besides

associating the corresponding learning objects

(annotated in the contents-repository ontology) for

teaching the course matter. This separation allows a

better reusability of the knowledge (e.g. in those

cases in which we are in front of a course offered in

different degrees whose contents differ very little).

The domain ontology for a given course is imported

from the “generic” domain ontology (it only contains

class definition and relations, but it does not contain

individuals) and the contents-repository ontology,

both shown in Figure 1. This ontology will include

the concepts of the course (individual of the class

Concept) and the relations among them, and also it

will point to the learning objects that can be used to

learn them.

Now the sequencing ontology for the course

imports the “generic” sequencing ontology (it only

contains class definitions and relations, but again it

does not contain individuals) and the domain

ontology particular for this course in order to define

the possible learning trajectories that the students

can follow through the concepts belonging to the

specific course (individuals of class Concept).

The author of the contents is provided with some

useful utilities – a domain tool, a sequencing tool

and a repository management tool – that allow him

to create, delete and modify ontologies associated to

the courses included with the tool. He will also be

able to populate the ontologies and to manage them

in a user friendly way, with no need to having any

knowledge on the format of the storage chosen.

AN AUTHORING ARCHITECTURE FOR ANNOTATING EDUCATIONAL CONTENTS - Domain, Sequencing and

Content-Repository Ontologies

3.1 Domain Ontology

The domain ontology is graphically shown in Figure

2. Class Course represents the subjects created

within the tool. For instance, Multi-agent Systems,

Operating Systems and Software Engineering are

some individuals of this class. The class contains

two data type properties, namely courseName and

courseDescription, the title and a brief description of

the course, respectively, and one object property

cHasObjective that points to the objectives to reach

(class Objective) through the course.

Concept

consistOf

isPartOf [0..1]

similarTo

oppositeOf

hasRequisite

isPrerequisiteFor

nameConcept : string

hierarchicalLevel : int

LearningObject

isDescribedBy describesTo

courseName : string

courseDescription : string

Course

description : string

Objective

cHasObjective

concHasObjective

achieveWith

Concept

consistOf

isPartOf [0..1]

similarTo

oppositeOf

hasRequisite

isPrerequisiteFor

nameConcept : string

hierarchicalLevel : int

LearningObject

isDescribedBy describesTo

courseName : string

courseDescription : string

Course

courseName : string

courseDescription : string

Course

description : string

Objective

cHasObjective

concHasObjective

achieveWith

Figure 2: Domain ontology.

The concepts constitute the knowledge of the

domain under study and they are collected in class

Concept. This class contains two data type

properties, nameConcept to identify the concept and

hierarchicalLevel to represent the level where the

concept is found in the concepts hierarchy. Notice

that level 0 is assigned to the root concept. There are

more object properties introduced later on to

describe the possible relationships among the

domain concepts.

Property consistOf is aimed to define a concept

hierarchy, and therefore, to establish a relationship

among a concept and its sub-concepts (e.g. we are

able to define chapters, sections, subsections and

terms which are under subsections), until reaching

an atomic concept which – from the point of view of

the teacher – does not need to be decomposed any

more. A concept must have at least a parent concept

(property isPartOf with maximal cardinality

restriction). Properties similarTo and oppositeOf

enable mapping a concept to other concepts that

have the same or different semantic meaning,

respectively.

In order to indicate concept restrictions through

which it is possible to advance/go back to/from a

given concept other properties are needed.

hasRequisite and isPrerequisiteFor (its inverse)

allow to point to concepts that must be known before

starting to study a determined concept, and the

concepts for which it is a prerequisite, respectively –

some conditions should be fulfilled to access the

study of the concepts.

Object property isDescribedBy (class Concept)

points to digital resources that explain a concept or

assess the knowledge stored about it. On the other

hand, object property describesTo is included in

LearningObject class to indicate which concepts a

learning object is related to.

3.2 Sequencing Ontology

The sequencing ontology defines the possible

learning paths that can be followed to learn the

concepts defined in the domain ontology. The

sequencing ontology imports the domain ontology to

add to class Concept (defined in the domain

ontology as well) the object properties shown in

Figure 3 to express how the teaching of the concepts

is sequenced. Object property sequence (functional)

points to the next concept to be followed; alternate

allows defining that it is possible to continue with

any one of the pointed concepts, whereas

xorAlternate only permits advancing through one of

the related concepts. For instance, Figure 3 shows

that a student starts learning concept C1, then C2

and later C3 (sequence). After this he could choose

to advance independently to any of the concepts

related to C3 through the alternative property (C4,

C5 and C13). If he arrives at concept C8 he will

have to continue to only one concept (C9 or C10).

Concept

sequence [0..1]

alternate

xorAlternate

Concept

sequence [0..1]

alternate

xorAlternate

C1 C2

C6 C7

C8 C9

C10

C11 C12

C13

C14

xor Alternate

x o r A l t e r n a t e

a l t e r n a t e

C1 C2

C6 C7

C8 C9

C10

C11 C12

C13

C14

xor Alternate

x o r A l t e r n a t e

a l t e r n a t e

Figure 3: Sequencing ontology and example.

ICEIS 2007 - International Conference on Enterprise Information Systems

3.3 Contents-repository Ontology

The contents-repository ontology includes metadata

to describe learning objects and their relationships.

To determine the grade of granularity of a learning

object is a fundamental decision in any project. The

degree of reusability of a learning object is largely a

function of its granularity, which is related with the

size of an object (the lower the size of an object, the

more reusable it will become) (South & Monson,

2000). Remember that in our approach the resources

that have a very low granularity are precisely the

learning objects (at the level of paragraph, image,

table, diagram, and so on). Thus, if there are learning

objects available at these levels, in every moment

any e-learning system is able to add/remove contents

at this level and to produce tailor-made learning

materials according to the preferences of a student.

Also, this facilitates showing didactical materials in

those devices that have a screen of limited

dimensions (e.g. a PDA) in form of a sequence of

pages. In this approach, learning objects that have a

greater granularity are built from smaller granularity

ones. For instance, the course chapter’s section will

be created by mixing several little chunks.

3.3.1 Learning Objects Description

Metadata is used to describe the learning objects. In

this work, some elements of the IEEE LOM (LOM,

2002) we have chosen, as this is a standard

recognized internationally and many metadata

schemas are based in it (e.g. IMS and SCORM

metadata). On the other hand, the inspiration of our

approach to describe the device features to correctly

display learning objects comes from the elements of

the LOM technical category and from the FIPA

Device Ontology (FIPA, 2002).

Class LearningObject includes the metadata

collection that describes a digital resource. As you

may observe on Figure 4, (a) a learning object is

created by one or several authors (createdBy), (b) it

has a set of keywords that describe it (hasKeyword),

is written in a given language (language), (e) it

manages a brief description (description), (f) it

incorporates a type of interactivity - taking values

active, exhibition and mixed - (interactivityType),

and, (g) it possesses a level of difficulty - very easy,

easy, average, difficult, and very difficult -

(difficultyLevel). The type of interactivity

denominated as active applies for documents where

a student interacts and/or performs operations (for

example, simulations, exercises, test questions),

whereas exhibition is applied to documents whose

objective is that the student just gets the content (for

example, text, images, sound). Lastly, object

properties hasLearningStyle, requiresDevice and

developSkill are introduced in order to point to the

learning styles that are better adjusted to a learning

object, the device where it is more correctly

visualized, and the intellectual skill developed when

it is used, respectively.

LearningObject

location : string

language : string

descript ion : string

interactivityType : string

difficultyLevel : int

Author

author : string

Keyword

keyword : string

LearningStyle

activeReflective : int

sensingIntuitive : int

visualVerbal : int

sequentialGlobal : int

Device

h a s K e y w o r d

c r e a t e d B y

r e q u i r e s D e v i c e

h a s L e a r n i n g S t y l e

Ski ll

d e v e l o p S k i l l

Cogni ti ve

bloomLevel : string

LearningObject

location : string

language : string

descript ion : string

interactivityType : string

difficultyLevel : int

Author

author : string

Keyword

keyword : string

LearningStyle

activeReflective : int

sensingIntuitive : int

visualVerbal : int

sequentialGlobal : int

Device

h a s K e y w o r d

c r e a t e d B y

r e q u i r e s D e v i c e

h a s L e a r n i n g S t y l e

Ski ll

d e v e l o p S k i l l

Cogni ti ve

bloomLevel : string

Figure 4: Description of a learning object.

So far, several learning styles theory have

already been used in adaptive educational systems

(Felder and Silverman, Dunn and Dunn, Honey and

Mumford models, and so on). For a deeper reading,

please consult (Stash, Cristea & De Bra, 2004). In

the work proposed in this paper, the scheme to

distinguish the student’s learning style is the one

proposed by the Felder-Silverman Learning Style

Model (FSLSM) (Felder & Silverman, 1988). The

FSLSM model distinguishes four dichotomous

dimensions for learning styles: active/reflective,

sensing/intuitive, visual/verbal, and

sequential/global, which gives place to sixteen

learning styles combinations. This decision has been

taken for two reasons. First of all, this model

provides a questionnaire to establish the dominant

learning style of each student and its results may

easily be linked to e-learning systems. Secondly, this

model has been sufficiently validated in many other

adaptive environments (Shi et al., 2003; Hong &

Kinshuk, 2004; Capuano et al, 2005; Peña, Marzo &

De la Rosa, 2005). Therefore, it seems reasonable to

annotate learning objects according to Felder and

Silverman model for the purpose of choosing the

best learning objects adapted to the student’s

learning style. Object property hasLearningStyle

points to suitable learning styles for a learning

object. The class LearningStyle offers four

AN AUTHORING ARCHITECTURE FOR ANNOTATING EDUCATIONAL CONTENTS - Domain, Sequencing and

Content-Repository Ontologies

properties of type integer that correspond to the four

dimensions of the FSLSM.

Object property requiresDevice in class

LearningObject of the contents-repository ontology

points to the device necessary to display a learning

object in a most satisfactory way. The class Device

describes the device technology that should be used

for displaying the contents correctly and in a suitable

time. A device satisfies certain hardware (class

Hardware) and software (class Software)

requirements. With regard to hardware, we consider

the following features to achieve a good user

satisfaction: the computer CPU type (cpu), the

network connection required (networkConnection),

the necessary memory (object property hasMemory),

the user interface characteristics (object property

hasUI), the capability to receive audio input (audio-

input) or to produce audio output (audio-output), as

well as information on the video card (videoCard).

Property hasMemory allows pointing to the

description of the features related to memory – the

amount of memory that the user device should

incorporate to show the learning object (amount) and

in which unit this amount is expressed

(unitMemory). Now, property hasUI allows pointing

to the information that describes the user interface -

the width of the screen (width), the height of the

screen (height), the unit for the width and height

parameters (unit) and if a colour screen (color) is

necessary. Regarding the software, features such as

the minimum and the maximum version capable of

using the resource (minimumVersion and

maximumVersion, respectively) and the name that

identifies it (nameSoftware) are included. We

distinguish the browser (Browser), the operating

system (OperatingSystem) and the pluggins

(Pluggin) as kinds of software.

Moreover, the taxonomy of learning objectives

proposed by Benjamin S. Bloom (Bloom, 1956) is

selected to annotate the learning objects according to

the instructional pedagogic role that they play from a

cognitive perspective. This supposition has been

adopted because this taxonomy is easy to understand

and it has also been widely applied (Soldatova &

Mizoguchi, 2003; Buckley & Exton, 2003; Ullrich,

2004). Class Cognitive has object property

bloomLevel to represent the intellectual skill that the

student develops when using it: knowledge,

comprehension, application, analysis, synthesis and

evaluation, just as identified in the cognitive domain

of the Bloom Taxonomy.

To conclude, class LearningObject also includes

object properties isEquivalentTo and

complementsTo. The fist one is for knowing the

resources that have the same semantic meaning,

whereas the second allows reusing resources that the

author of the contents thinks that are necessary to be

grouped together in a course. For instance, a

diagram should be grouped together with the

paragraph and/or audio that describes it; a simulation

should be reused together with the paragraphs that

explain a concept, etc.

3.3.2 Types of Learning Objects

The individuals of class LearningObject may be

theoretical explanations, practical explanations, or

evaluation questionaires belonging to classes

TheoreticalContentObject, PracticalExplanation and

IndividualEvaluation, respectively. Notice that these

classes are subclasses of class LearningObject. They

are related to each others, as represented

schematically in Figure 5 through rTCO_IE,

rTCO_PE and rIE_PE. Let us highlight that in the

ontology, instead of rTCO_PE there is an object

property to indicate that a theoretical content object

can be related with several practical explanations.

Another object property expresses that a practical

explanation can be related with several theoretical

content objects. Exactly the same idea is applicable

to rTCO_IE and rIE_PE.

Classes TheoreticalExplanation and

PracticalExplanation represent the theoretical and

practical explanations, respectively, that are shown

to the students (see Figure 5). In order to compose

the theoretical explanations several types of formats

are proposed (classes Text, Audio, Video, and

Image). This way, the theoretical explanations that

appear to the verbal students are formed by text

and/or audio, whereas videos and images are shown

to the more visual students. Classes Text, Audio,

Video and Image are subclasses of Theory, which

includes the data type property order to describe

how to link the theoretical contents. Class Text has

the property role that allows expressing whether it is

a text chunk associated to a definition, summary,

law, theorem, or proof, among others; whereas class

Image has the property type to describe that it is a

graph, figure, graphical scheme, etc. In order to

realize practical explanations, examples, simulations

and animations (classes Example, Simulation,

Animation) can be used.

ICEIS 2007 - International Conference on Enterprise Information Systems

LearningObject

isa

TheoreticalContentObject IndividualEvaluation

PracticalExplanation

Video Audio Text Image

Exercise TestQuestion Simulation Animation Example

NumericSolution

FreeResponseAssociation

IncompletePhrase

SingleSelection

TrueFalse

MultipleSelection

isEquivalentTo

complementsTo

isa

isa isaisaisa

isa

isa isa isaisa isa

isa

rTCO_IE rIE_PE

rTCO_PE

Theory

order : int

hasTheory

type : string

role : string

hasLear ningStyle

LearningStyle

LearningObject

isa

TheoreticalContentObject IndividualEvaluation

PracticalExplanation

Video Audio Text Image

Exercise TestQuestion Simulation Animation Example

NumericSolution

FreeResponseAssociation

IncompletePhrase

SingleSelection

TrueFalse

MultipleSelection

isEquivalentTo

complementsTo

isa

isa isaisaisa

isa

isa isa isaisa isa

isa

rTCO_IE rIE_PE

rTCO_PE

Theory

order : int

hasTheory

type : string

role : string

hasLear ningStyle

LearningStyle

Figure 5: Types of learning objects.

The individuals of class IndividualEvaluation are

used to evaluate the knowledge acquired by the

student. Class IndividualEvaluation has two

subclasses (Exercise and TestQuestion) that contain

the exercises and the test questionnaires,

respectively, that a student has to solve. Class

Exercise contains four subclasses associated to

different kinds of statements in which (1) a question

is posed where the student has to answer with a

numerical solution (NumericSolution), (2) it is

necessary to complete in one o more points with a

phrase, a word or a cipher (IncompletePhrase), (3)

the student has to answer with one or several

paragraphs (FreeResponse), and, (4) he must

establish the relations between the elements of two

columns (Association). On the other hand, class

TestQuestion has several subclasses to highlight the

different types of test questionnaires that are shown

to the student. For instance, a student has to choose

between one of two alternatives (TrueFalse), one of

three or more alternatives (SingleSelection), or all

the correct ones from a series of alternatives

(MultipleSelection).

4 CONCLUSIONS

In this paper a proposal for an authoring architecture

based in several ontologies, namely domain

ontology, sequencing ontology and contents-

repository ontology, has been introduced. To

annotate a learning object, characteristics such as the

most appropriate learning style to a student, the

device used to access the contents and the skill that

is expected to be developed when the student

approaches the contents are considered in the

contents-repository ontology. In the domain

ontology the concepts and the relations between

them we describe, just as appearing in the course

domain. In the sequencing ontology the possible

learning paths that can appear through the concepts

defined in the domain ontology we define.

As future work, it would be important to provide

a more sophisticated learning objects assembly

mechanism allowing to generate them according to

formats generally used at present (IMS and

SCORM) and fitted to the profile of each student.

Moreover, we are engaged in using these objects in

an agent-based intelligent tutoring system proposed

very recently (Fernández-Caballero et al., 2006) in

order to improve the proposal’s adaptivity

capacities.

ACKNOWLEDGEMENTS

This work is supported in part by the Spanish

CICYT TIN2004-08000-C03-01 and the Junta de

Comunidades de Castilla-La Mancha PAI06-0093

grants. José M. Gascueña is the recipient of a

Predoctoral Scholarship awarded by Junta de

Comunidades de Castilla-La Mancha under grant

PBI06-0099.

AN AUTHORING ARCHITECTURE FOR ANNOTATING EDUCATIONAL CONTENTS - Domain, Sequencing and

Content-Repository Ontologies

REFERENCES

Antoniou, G. & van Harmelen, F. (2004). A Semantic

Web Primer. The MIT Press, Massachussets Institute

of Technology. Cambridge, Massachussets.

Baloian, N.A., Galdames, P., Collazos, C.A. & Guerrero,

L.A. (2004). A model for a collaborative recommender

system for multimedia learning material. Lecture

Notes in Computer Science, 3198, pp. 281-288

Bloom, B.S. (Ed.). (1956). Taxonomy of educational

objectives: The classification of educational goals:

Handbook I, Cognitive Domain. New York, Toronto:

Longmans, Green, 1956.

Buckley, J. & Exton Ch. (2003). Blooms' taxonomy: A

framework for assessing programmers' knowledge of

software systems. Proceedings of the 11th IEEE

International Workshop on Program Comprehension,

IWPC'03, pp. 165-174.

Capuano, N., Gaeta, M., Micarelli, A., & Sangineto, E.

(2005). Automatic student personalization in preferred

learning categories. In 11th Conference International

on Human Computer Interaction, HCI International.

Dean, M. & Schreiber, G. eds. Bechhofer, S., van

Harmelen, F., Hendler, J., Horrocks, I., McGuinness,

D.L., Patel-Schneider, P.F. & Stein, L.A. (2004).

OWL Web Ontology Language Reference. W3C

Recommendation 10 February 2004.

http://www.w3.org/TR/owl-ref/

Duitama Muñoz, J.F. (2005). Un modèle de composants

pédagogiques pour des cours adaptatifs sur le Web.

Ph. D. Thesis, Institut National des

Télécommunications, France. http://etna.int-

evry.fr/~duitama/thesis-home/PhD-Thesis-05-04-

2005pdf.zip.

Felder, R.M, Silverman, L.K. (1988.) Learning and

teaching styles in engineering education. Journal of

Engineering Education, 78(7), pp. 674-681. Retrieved

October 25, 2006, from http://www.ncsu.edu/felder-

public/Papers/LS-1988.pdf

Fernández-Caballero, A., Gascueña, J.M., Botella, F.,

Lazcorreta, E., 2006. Distance learning by intelligent

tutoring system. Agent based architecture for user-

centred adaptivity. In Enterprise Information Systems

VII.

FIPA (2002). FIPA Device Ontology Specification. 6-12-

2002. Retrieved October 25, 2006, from

http://www.fipa.org/specs/fipa00091/SI00091E.pdf

Gruber, T. R. (1993). A translation approach to portable

ontology specifications. Knowledge Acquisition, 5(2),

pp. 199-220.

Hong H., Kinshuk. (2004). Adaptation to student learning

styles in web based educational systems. Proceedings

of ED-MEDIA 2004, World Conference on

Educational Multimedia, Hypermedia &

Telecommunications, pp. 491-496.

Jovanovic, J., Gasevic, D. & Devedzic, V. (2006).

Ontology-based automatic annotation of learning

content. International Journal on Semantic Web and

Information Systems, 2(2), pp. 91-119.

Leidig, T. (2001). L3—towards an open learning

environment. ACM Journal on Educational Resources

in Computing, 1(1), article #5, 11 pages.

LOM (2002). IEEE 1484.12.1-2002, Draft Standard for

Learning Object Metadata, (2002).

McGreal R. (2004). Learning objects: A practical

definition. International Journal of Instructional

Technology and Distance Learning, 1(9), pp. 21-32.

Peña, C.I., Marzo, J.L., & De la Rosa, J.L. (2005).

Intelligent agents to improve adaptivity in a web-based

learning environment. In C. Ghaoui, M. Jain, V.

Bannore & L.C. Jain (Eds.), Knowledge-Based Virtual

Education, Springer-Verlag, pp. 141-170

Ronchetti, M. Saini, P. (2004). Knowledge management in

an e-learning system. Proceedings of the 4th IEEE

International Conference on Advanced Learning

Technologies, ICALT 2004, pp. 365-369.

Santacruz-Valencia, L., Navarro, A., Aedo, I. & Kloos C.

(2005). An ontology-based mechanism for assembling

learning objects. Proceedings of E-Learning on

Telecommunications, ELETE 2005, pp. 472-477.

Shi, H., Rodriguez, O., Shang, Y. & Chen, S. (2003).

Integrating adaptive and intelligent techniques into a

web-based environment for active learning. In Ed: C.

T. Leondes (Ed.), Intelligent Systems: Technology and

Applications. CRC Press/Lewis Publishers.

Shimic, G., Gasevic, D., & Devedzic, V. (2006).

Classroom for the Semantic Web. In M.D. Lytras, A.

Naeve (Eds.), Intelligent Learning Infrastructure for

Knowledge Intensive Organizations. Information

Science Publishing, pp. 251-283.

Silva Muñoz, L. & Palazzo M. de Oliveira, J. (2004).

Adaptive web-based courseware development using

metadata standards and ontologies. Proceedings of the

16th International Conference on Advanced

Information Systems Engineering, CAISE 2004.

Lecture Notes in Computer Science, 3084, pp. 414-

428.

Soldatova, L., & Mizoguchi, R. (2003). Ontology of tests.

Proceedings of Computers and Advanced Technology

in Education, pp.175-180.

South, J.B. & Monson, D.W. (2000). A university-wide

system for creating, capturing, and delivering learning

objects. Section 4.2 of the book The Instructional Use

of Learning Objects: Online version

http://www.reusability.org/read/chapters/south.doc

Stash, N., Cristea, A., & De Bra, P. (2004). Authoring of

learning styles in adaptive hypermedia: Problems and

solutions. Proceedings of the 13th International World

Wide Web Conference, pp. 114-123.

Ullrich, C. (2004). Description of an instructional

ontology and its application in web services for

education. Proceedings of the Workshop on

Applications of Semantic Web Technologies for E-

learning, SW-EL'04, pp. 17-23.

Wiley, D.A. (2001). Connecting learning objects to

instructional design theory: A definition, a metaphor,

and a taxonomy. In D.A. Wiley (Ed.), The

Instructional Use of Learning Objects: Online Version.

ICEIS 2007 - International Conference on Enterprise Information Systems