SALO
Sharable Auto-adaptive Learning Object
Ignacio Gutiérrez Menéndez, Mª del Puerto Paule Ruiz and Juan Ramón Pérez Pérez
Universidad de Oviedo, Dpto. Computer Science, Oviedo, Spain
Keywords: e-Learning, Dynamic languages, Adaptation.
Abstract: New pervasive computing scenarios such as ubiquitous learning (u-learning) and mobile learning (m-
learning) demands real time adaptation. Getting such adaptation couldn’t be possible with the current
technologies. It is necessary a change in the development of pervasive e-learning systems using dynamic
technologies and including them in both: platforms and e-learning content specifications. In this paper we
define and develop the concept of Sharable Auto-Adaptive Learning Object (SALO) like a Learning Object
which includes content and describe its behaviour thanks to dynamic languages. Such features allow it to
change and include new resources and behaviour, in a dynamic way and using the user’s context at 'anytime,
anywhere, from any device'.
1 INTRODUCTION
The reuse of electronic resources is a complex
matter that should guarantee more attractive,
effective, efficient and accessible learning
experiences for students (Campbell, 2003).
Ubiquitous learning allows students to learn in any
place and moment and with any device (Abarca et
al., 2006). A context-aware approach is needed
because each learner has unique needs and context,
but also provides focused and structured learning
(Farrell et al., 2004).
The interoperability in the educational content
has allowed the creation of the learning objects
(IEEE, 2002) (Willey, 2000). The use of open
standards in course construction has brought about
Content specifications such as Sharable Content
Object Reference Model (SCORM;
http://www.adlnet.gov), IMS Learning Design (IMS-
LD; http://www.imsglobal.org) and IMS Common
Cartridge (IMS-CC; http://www.imsglobal.org/cc/).
The specifications work with an object model
that allows serializing the learning objects,
independently of the tool or application which they
have been created or processed. However, the actual
specifications present, under our point of view, a
limitation: the impossibility of storing or serializing
the objects behaviour. Stored and distributed
learning objects are just content containers without
any own behaviour. Actual Learning Management
System (LMS) must provide the behaviour.
From our point of view, this way of
implementing the behaviour makes impossible an
adaptation in real time, restricting the creation of
ubiquity learning (Weiser, 1993), mobile learning
and context-aware learning systems. Our proposal
follows changing this approach. The aim is to
generate learning objects with data and behaviour.
This idea is not new, in (Bailey et al., 2002) appears
the concept of the Fundamental Open Hypermedia
Model (FOHM) and in (Zouaq et al., 2008) presents
an ontology-based approach for the dynamic
generation of learning knowledge objects (LKO).
We have developed and implemented a SALO’s
(Sharable Auto-adaptive Learning Objects), objects
which are fully autonomous, where both content and
behaviour are stored. In order to achieve this
approximation, it is necessary to use dynamic
adaptation (Yang et al., 2002), which allows the
possibility of adapting the learning objects to the
context in which they are being executed.
2 RELATED WORK
2.1 Adaptive Educational Systems
In general, the Adaptive Educational Hypermedia
Systems (Brusilovsky, 2001) are based on the use of
Adaptive hypermedia techniques (Brusilovsky,
387
Gutiérrez Menéndez I., del Puerto Paule Ruiz M. and Ramón Pérez Pérez J..
SALO - Sharable Auto-adaptive Learning Object.
DOI: 10.5220/0003303103870390
In Proceedings of the 7th International Conference on Web Information Systems and Technologies (WEBIST-2011), pages 387-390
ISBN: 978-989-8425-51-5
Copyright
c
2011 SCITEPRESS (Science and Technology Publications, Lda.)
1996) to solve a number of the problems associated
with the use of educational hypermedia. Different
ways of implementing the adaptation has been used
in educational systems (Paule et al., 2008):
1. Use of rules that activate one another when
certain actions happen, these rules can also create
new rules, e.g.: AHA! (http://aha.win.tue.nl) and
MOT
(http://www.dcs.warwick.ac.uk/~acristea/mot.html)
are both systems that belong in this first category.
2. The second kind is the creation of proprietary
objects with a particular format created for each
application. AuldLinky (Michaelides et al., 2001) is
a system that belongs to this category.
3. In (Cristea et al., 2007) the authors describe two
new representation languages that emerged in the
process of adaptation: a common format for defining
the static material, CAF, and an extended adaptation
language for the description of the dynamic
behaviour, LAG.
In general, the use of rules is important to support
the adaptive capabilities of the system; however, the
adaptive applications generate a content that cannot
be reused by other systems. Regarding the creation
of proprietary objects for each application, sharing
these objects among different applications is
difficult since these objects do not follow any
standard format. The use of LAG is a step forward in
this field, although it is not compatible with the
existing specifications on the market.
2.2 Adaptation in the e-Learning
Content Specifications
Merging adaptive hypermedia with e-learning seeks
to adapt e-learning information according to the user
needs and context. The specifications use the
sequencing in order to get the adaptation, SCORM
has the SN (Sequencing & Navigation) which by
means of rules and depending on user model it
allows the LMS to show the right content. The
decision taken by the sequencing is done by a static
engine and a number of rules that are executed
according to some data returned by the learning
object.
The learning objects are mere data containers,
but without any related behaviour, leaving this
behaviour to the LMS, so developers need to build a
set of sequencing rules and an API defined by the
content specification. It would be a great advance to
provide learning objects with a complex and
independent behaviour from the platform. With this
goal, LMS just needs to care about the content
administration task, releasing it from the task of
managing object relationships.
3 OUR PROPOSAL:
SHARABLE AUTO-ADAPTIVE
LEARNING OBJECTS
We consider that an auto-adaptive learning object
must have the following features:
1. Own content according to the learners’ needs and
context
2. Own behaviour allowing:
- To be sufficiently aware of the learners’ context
- To be able to take full advantage of pervasive
computing
3. To be reused many times independent of
software and platforms versions
4. To be interoperable; compatible with the content
specifications and with current LMSs
The first three features require dynamic adaptation.
In order to get dynamic adaptation, we need to use
reflective and reflection techniques (Pattie, 1987)
and those techniques are offered currently by
dynamic programming languages (Ortín et al.,
2005). From our point of view, it is necessary the
addition of dynamic features to a LMS to get
context-aware, ubiquity and multi-mode e-learning
systems. These features implies many changes in
design and development of actual learning
platforms, however these changes are compatible
with the actual e-learning frameworks such as
OKI(http://www.okiproject.org), ELF
(http://www.elframework.org) or IMS Abstract
Framework(http://www.imsglobal.org/af/), because
these frameworks define a set of services which are
independent of the programming language.
3.1 Our Proposal:
Sharable Auto-adaptive
Learning Object (SALO)
We define as a Shareable Auto-Adaptive Learning
Object (SALO) as a learning object which is able to
describe its own behaviour, being independent from
the LMS, adapting itself to the context and being
reusable for other e-learning systems.
A SALO contains data and behaviour (Figure 1),
packed both of them with SCORM specification,
which makes it independent of the LMS.
WEBIST 2011 - 7th International Conference on Web Information Systems and Technologies
388
LearningObject
Content
<organization>
<item>
<item>
<item>
<item>
SALO
Content
Behaviour
SALO
Content
Behaviour
SALO
Content
Behaviour
Sequencing
Currentspecifications Proposalofanewspecification
LearningObject
Figure 1: Transition from LO to SALO.
The creation process of a SALO is formed by an
editing process plus a run time process and it is
divided into four parts:
- To choose the content: pdf, html, or any kind of
educational resource.
- Addition of behaviour: It consists on a set of
techniques and methods of adaptive hypermedia and
behaviour rules, which define the instructional
design of the LO and they are implemented in Ruby.
Teacher has an edit tool in order to include such
techniques and rules into a SALO.
- Packing: SALO’s learning content is packed in a
similar way as SCORM does, but also our process
packs its behaviour. Later on, content and behaviour
will be used by the dynamic engine inside a LMS to
make the user adaptation possible.
- Adaptive environment: It is done by the LMS and
the adaptive engine. The adaptive engine loads the
content and behaviour of the LO in memory, starting
its execution. The engine also processes the SALO´s
learning content and returns it to the LMS. Once the
interaction between SALO and the user has finished,
a terminate request is gotten and according to the
object’s behaviour the next SALO is selected and
the process begins again. If the user decides to quit
the course, the state is stored in the system.
The engine has two main functions: load and run the
SALO. Firstly the loading process connects to a
repository and it will take every data and resources
building a current LO. Also, it will take the
behaviour which will be transformed, using
reflective techniques, into methods that will be
attached to the LO previously built producing a
SALO in memory. Secondly, the SALO is executed
and it uses its new methods to change its view to the
user.
The previously described process adapts the
SALO to the user at runtime and can be modified
just adding/removal behaviour in the repository. At
this point it is interesting to highlight that every
addition we have done to the SCORM specification,
to create a SALO, is compatible with the current
SCORM format, because, this extra information will
be ignored by the current LMSs.
3.1.1 Preliminary Results
We have tested the implementation of a SALO with
20 learners of University of Oviedo. The goal of this
test is to show that the adaptation technologies
proposed is feasible.
For the test every user had to complete two
courses - with and without adaptation - and they
must be done twice each one. The subjects chosen
for the courses have been related to “golf” and
“basics of programming”.
The results, which are not conclusive, shows that
most learners appreciate the combination of
adaptation technologies adopted and the support
offered by the implementation because they notice
as significant visual changes between the adapted
course and the course without adaptation. These
results go in parallel with other similar studies done
by other researchers in the Adaptive Hypermedia
Educational Systems.
4 CONCLUSIONS
AND FUTURE WORK
Nowadays, researches related to ubiquity and
context-aware e-learning systems are in great
expansion. These systems need adaptation to the
current user context in real time. Dynamically typed
programming languages are able to include dynamic
adaptation to the e-learning content, as it has been
shown in this paper, however using dynamic
technologies means a change in design, develop and
implementation of the current LMS´s.
A SALO changes the meaning of the current
LO’s, giving them behaviour and making them
independent of the LMS´s where they are being run.
They are compatible with current content
specifications like SCORM, and also with LMs´s
like Moodle. Nowadays there are LMS´s
implemented with dynamic languages like Moodle,
implemented in PHP. This kind of implementation
of LMS makes easier to add the adaptation engine
which allows the SALO’s execution.
Currently we are exploring the possibility of
creating new adaptive strategies in runtime
according to a user model and how to do the
SALO - Sharable Auto-adaptive Learning Object
389
connection between the user model and its
translation in adaptive rules.
ACKNOWLEDGEMENTS
This work has been partially funded by the
Department of Science and Technology (Spain)
under the National Program for Research,
Development and Innovation; project TIN2008-
00276.
REFERENCES
Abarca, M. G., Alarcon, R. A., Barria, R., Fuller,
D.(2006): Context-Based e-Learning Composition and
Adaptation, Mapping OWL-DL into ORM/RIDL,
Herrero et al. (Eds.): OTM Workshops, LNCS 4278,
pp. 1976 – 1985.
Bailey, C., Hall, W., Millard, D. E. & Weal, M. J. (2002):
Towards Open Adaptive Hypermedia. In Proceedings
of 2
nd
International Conference Adaptive Hypermedia
and Adaptive Web-Based Systems, LNCS 2347, pp 36-
46.
Brusilovsky, P. (1996). Methods and techniques of
adaptive hypermedia. User Modelling and User
Adapted Interaction, 6 (23), 87-129.
Brusilovsky, P. (2001). Adaptive Educational
Hypermedia, Proceedings of the Tenth International
PEG Conference, Digital Media Institute, Tampere
University of Technology, Tampere, Finland.
Campbell, L. M. (2003) Engaging with the learning object
economy. In A. Littlejohn (Ed.), Reusing online
resources: A sustainable approach to e-Learning,
London: Kogan Page.
Cristea, A., Smits. D., De Bra, P. (2007), Towards a
generic adaptive hypermedia platform: a conversion
case study. Journal of Digital Information (JoDI),
Special Issue on Personalisation of Computing &
Services, Vol 8, No 3.
Farrell, R. G., Liburd, S. D., Thomas, J.C.(2004),
Dynamic assembly of learning objects, Proceedings of
the 13th international World Wide Web conference,
pp. 162-169.
IEEE. (2002), Learning Technology Standards Committee
(PDF), Draft Standard for Learning Object Metadata
(2002). IEEE Standard 1484.12.1, New York: Institute
of Electrical and Electronics Engineers. Retrieved
Sept 1, 2010, from http://ltsc.ieee.org/wg12/files/
LOM_1484_12_1_v1_Final_Draft.pdf.
Michaelides, D. T., Millard, D. E., Weal, M. J. and De
Roure, D., C. (2001) Auld Linky: A Contextual Open
Hypermedia Link Server, In Proceedings of the 7th
Workshop on Open Hypermedia Systems, ACM
Hypertext 2001 Conference Aarhus, Denmark.
Ortín, F., Díez, D (2005) Designing an Adaptable
Heterogeneous Abstract Machine by means of
Reflection. Elsevier Information and Software
Technology, 47 (2), 81-94.
Pattie Maes. (1987). Computational Reflection. PhD.
Thesis. Laboratory for Inteligence Artificial. Vrije
Universiteit Brussel.
Paule, M. P., Fernández, M. J., Ortín, F., Pérez, J. R.
(2008). Adaptation in current e-learning systems.
Elsevier Computer Standards and Interfaces, 30 (1-2)
62-70.
Weiser, M. (1993), Some Computer Science Issue in
Ubiquitous Computing. Communications of the ACM.
36(7) 74-84.
Wiley, D. A. (2000) Connecting learning objects to
instructional design theory: A definition, a metaphor,
and a taxonomy. En D. A. Wiley (Ed.), The
Instructional Use of Learning Objects. Retrieved Sept
1, 2010, from http://reusability.org/read/chapters/
wiley.doc.
Yang, Z., Cheng, B. H. C., Stirewalt R. E. K, Sowell, J., S.
Sadjadi, M., McKinley, P. K. (2002), An Aspect
Oriented Approach to Dynamic Adaptation WOSS
’02, Charleston, SC, USA.
Zouaq, A., Nkambou, R., Frasson, C. (2008): Bridging the
Gap between ITS and eLearning: Towards Learning
Knowledge Objects. Proc. Ninth Int’l Conf. Intelligent
Tutoring Systems (ITS ’08), pp. 448-457.
WEBIST 2011 - 7th International Conference on Web Information Systems and Technologies
390