A New Approach Based on Learning Services to Generate

Appropriate Learning Paths

Chaker Ben Mahmoud

, Fathia Bettahar

, Marie-Hélène Abel

and Faïez Gargouri

IResCoMath, University of Gabès, Gabès, Tunisia

Université de Technologie de Compiègne, Compiègne, France

ISIM’Sfax, University of Sfax, Sfax, Tunisia

Keywords: e-Learning, Learning Paths, Learning Service, Ontology, Composition, OWL-S.

Abstract: This article presents a new approach to provide learners learning paths adapted to their profiles. These

courses are generated as the automatic composition of learning services. It is made up of three modules:

search module, matching module and composition module. Our approach is based on new model of learning

service (SWAP) that extends semantic web service (OWL-S) to describe the semantics of learning modules

and facilitated the discovery of learning paths adapted to each learner.

1 INTRODUCTION

With the advance in technology, the mode of

education has been changing its traditional method

of teaching and learning toward online learning,

which can be defined as the acquisition, use and

dissemination of knowledge provided primarily by

electronic media (Sultana, 2010).

e-Learning is a type of education that offers

advantages to traditional learning in terms of

independence. In fact, learners can study from

anywhere at any time and communicate with the e-

learning system or other learners by e-mail,

electronic forums, chat, video and other forms of

communication based on Web (Zhen, 2009).

It is not only characterized by easy access to

learning resources (anytime and anywhere), but it

also supports other functionalities such as the

definition of personal learning goals, and

communication and synchronous or asynchronous

collaboration between learners and trainers. In this

context, objects and learning resources and their

providers play a very important role in learning

applications (Gang, 2004). This content, which is

created by educators and used by learners is usually

handled, stored and exchanged in Learning object

units (LOs).

Learning objects (LOs) are teaching units

(studies) designed to meet specific learning

objectives. These objects are modeled in terms of

content items, practice items and evaluation

elements to cover a subject of learning and an

assessment. They represent reusable granules that

can be drawn independently from the delivery

medium and dynamically access the network (Zhen,

2009).

As the number of resources and suppliers grows,

it becomes very difficult for the learner to search the

needed resources. A first solution is to add metadata

to the specification of the learning object.

At this stage, several standardization efforts have

been launched including LOM and SCORM. They

present a set of specifications adapted from multiple

sources of learning.

These standard descriptions of learning resources

are focused on the characterization of content rather

than on their use. In fact, they have limitations in the

context of heterogeneous learning objects. In

addition, the definition of specialized courses

according to desired skills requires a composition of

learning objects to provide the learner a personalized

learning course.

These problems of interoperability, reusability

and composition of learning content can be solved

by using the principles of Web service paradigm.

Web services are defined as open standards that

provide a flexible solution for integrating

heterogeneous and dynamic applications that enable

interoperability between different systems (Ngamnij,

2005).

643

Ben Mahmoud C., Bettahar F., Abel M. and Gargouri F..

A New Approach Based on Learning Services to Generate Appropriate Learning Paths.

DOI: 10.5220/0004940006430646

In Proceedings of the 16th International Conference on Enterprise Information Systems (ICEIS-2014), pages 643-646

ISBN: 978-989-758-028-4

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

The rest of the paper is organized as follows:

Section II presents an overview of approaches using

learning services. Section III details a new approach

to provide learners learning paths adapted to their

profiles. This approach consists of three components

or layers: Presentation layer, Business layer and

Model layer. Section IV concludes the paper and

outlines future works.

2 RELATED WORK

e-Learning and the integration of learning resources

are topics of growing interest in recent years.

Zniber (ZNIBER, 2010) presented an approach

to building personalized pathways called POPS

(Process-Oriented Pedagogic Service) by composing

services dynamically. This approach is a conceptual

framework that defines a model for describing the

pedagogical services. This model of Pedagogical

Service provides a set of concepts to describe the

services. According to Zniber, a pedagogical service

is composed of three parts: "profile”, "structure" and

"behavior". The "Profile" part describes the general

appearance of the pedagogical service.

It corresponds to the service interface and will be

used when searching for a match between the

available services and the learners’ intentions. It is

composed of the definition of a pedagogical

objective and a learning context. The "structure" part

describes the organization of the process to achieve

the pedagogical objective. It is defined by a process

and an initial position and a final position. The

"behavior" part is the "executable" level of service.

It describes the use of the service by a learner and it

takes the form of an implementation plan with

activities and resources to be mobilized. The

behavior part is composed of resources and links to

use these resources.

In this approach, the author has used ontologies

to describe both pedagogical services and make the

intentions of learners who need personalized

pathways. This description of pedagogical services

is based on two ontologies: one to describe

pedagogy and another to describe the teached

domain. These ontologies are used to associate a

semantic description of the elements of services.

D'Mello (D’Mello, 2012) proposed a functional

and semantic approach to describe the e-learning

web services with various objects and learning

resources. It defines a learning service as an

extension of the WSDL2.0 document structure

integrating functional semantics to e-learning web

services and their operations in order to ensure the

publication of these services. This extension is

composed of a set of new elements. The element

"documentation" is chosen to include the

information necessary for better service discovery in

WSDL. The label entitled "operationDesc" is

defined to insert the functional semantics of all

operations present in the learning service. The new

elements "operationList", "operation", "action",

"call", "object" and "name" are used within the

element “operationDesc”. These new elements are

defined in the XML schema that governs the

structure of the extended documentation element.

Functional semantics of a transaction are defined

inside the element “semantics”, the element is

placed inside of the element “operation”. Elements

such as “action”, “qualify”, “object” and “not” are

used in the element “semantics” that provides a

functional description of a learning operation.

Figure 1: Example of a Web service learning.

3 APPROACH BASED

LEARNING SERVICES

We propose an approach that allows us to provide

learners with learning paths adapted to their profiles.

These courses are compositions learning services.

This approach consists of three components or

layers:

• Presentation layer: it provides the learner with an

interface to formulate queries.

• Business layer: is the functional part which

processes the learner’s queries by searching and

composing pedagogical services

< ?xml version="1.0" encoding="utf-8" ?>

<operationDesc

xmlns :xf="http://www.w3.org/2011/XMLSchema-Instance"

xf:schemaLocation="http://www.sejc.ac.in/csedept/desSchema/desc.xsd"

xmlns =" http://www.sejc.ac.in/csedept/desSchema" >

<operationName> Get Course Material

</operationName>

</semantics>

</operation>

</operationList>

</operationDesc >

</documentation>

</description>

ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems

644

• Model layer: represents the basis for existing

pedagogical services and the used ontologies

(both teached domain ontology and objectives

ontology).

Our approach is described by three modules

including: search module, matching module and

composition module.

The learner expresses these learning needs as a

query (search module). The system takes the

learning query and verifies the correspondence

between the request of the learner and the learning

web service model (SWAP) in order to eliminate

any conflict of information presentations (matching

module). Once the request is approved, the system

starts the search of appropriate learning web services

from a SWAP database. This search may lead to a

composition of services to satisfy the learner

(composition module).

The result is a personalized path. At this stage,

the learner can give its opinion on the proposed path,

if the opinion is positive, the SWAP database is fed

by the composed service.

Figure 2: Architecture-based learning Web services.

3.1 Presentation Layer

In this layer, we define the profile of the learner and

these learning needs in the form of learning requests.

The learner profiles are used to provide the learning

system with relevant information to adapt learning to

the knowledge, skills, characteristics, preferences

and learning goals of the learner. In our approach,

we define a profile of learner as the learning level

and all learning services consumed by the learner.

This profile will help us to provide the best and the

most appropriate learning service suited to learners.

3.2 Business Layer

At this layer, we proceed the matching between the

request and the model of learning service. This step

can correctly interpret the semantics of the request

of learner and improve the accuracy of responses. In

this context, this matching can offer better speed of

execution from a first simple filtering of learning

services that corresponds to check if the learning

request corresponds to a simple learning service.

After the first filtering step, we proceed to the

composition of a set of learning services to satisfy

the learner request. The result of this composition is

a personalized path in the form of an oriented graph.

3.3 Model Layer

This layer contains a basis of learning services and

the used ontology. These ontologies are ontology of

objectives and ontology of learning.

 Model of a learning service: We present the

model of learning service (SWAP) as an

extension of semantic web service (OWL-S) to

describe the semantics of learning modules and

facilitated the discovery of learning paths

adapted to each learner.

This model is composed of two layers:

 Environment layer: that describes the basic

information of a learning service.

 Action layer: present the various existing

learning operations as well as interactions

between them represented in the form of oriented

graphs.

Figure 3: Model of a learning service.

ANewApproachBasedonLearningServicestoGenerateAppropriateLearningPaths

645

4 CONCLUSION

This paper is a study to implementing a new e-

learning approach that enables to provide learners

with learning paths adapted to their profiles. These

paths are compositions of learning services. At

present our work is at an early stage and we still

need to evaluate the approach. In the future, we will

proceed to detail the different layers of our

approach.

We relied on MEMORAe project conducted in

the laboratory Heudiasyc of the University of

Technology of Compiègne (UTC), whose goal is to

model and design a collaborative environment

fostering organizational learning and knowledge

management (Leblanc, 2008).

REFERENCES

Afroza Sultana and Ishrat Sultana, 2010, “E-School: A

Web-Service Oriented Resource Based Elearning

System”, In Proceedings of the International

Conference on Networking and Information

Technology, IEEE.

ZHU Zhen, 2009, “Design and Implementation of Web-

services based ELearning System”, In Proceedings of

the First International Workshop on Education

Technology and Computer Science, IEEE.

Zhao Gang, Yang ZongKai and Yan Kun, 2004, “Design

and Implementation of a Distributed Learning

Resource Registry System”, In Proceedings of the

Fourth International Conference on Computer and

Information Technology (CIT’04), IEEE.

Ngamnij Arch-int, Chidchanok Lursinsup and Peraphon

Sophatsathit, 2005, “A Reference Architecture for

interoperating existing e-Learning Systems using

Metadata and Web Services Model”, In Proceedings

of the International Conference on Intelligent Agents,

Web Technologies and Internet Commerce (CIMCA-

IAWTIC’05), IEEE.

Najlaa ZNIBER, 2010, “Modèle orienté service pour la

conception de parcours pédagogiques personnalisés”,

thèse, AIX-MARSEILLE UNIVERSITÉ.

Demian Antony D’Mello, 2012,“ Functional semantics

aware broker based architecture for e-learning web

services”, International Journal on Integrating

Technology in Education (IJITE) Vol.1, No.1,

December.

Adeline Leblanc, 2008, Marie-Hélène Abel, “E-

MEMORAe2.0: An e-learning Environment as

Learners Communities Support,” IJCSA 5(1), pp.108-

123.

ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems

646