e-Learning systems personalization - new requirements and solutions

Document Sample
e-Learning systems personalization - new requirements and solutions Powered By Docstoc
					    e-Learning systems personalization - new requirements and
                       ing. Borozan Ana-Maria, Bucharest, Romania,
                             ing. Trandafir Ileana, Bucharest, Romania,

The paper presents an IMS standards-based approach, adopted to specify the
personalization aspects throughout the system design phase, adopting innovative
solutions in 3 domains: teaching-learning process modelling, learner modelling and
content modelling.
Keywords: e-learning, personalization, IMS standards, modelling

1. Lifelong learning - new requirements for the e-learning systems
In the knowledge based society the lifelong learning becomes an essential process,
sustained mostly by the information and communication technologies. Many
analysts and researchers underline the fact that lifelong learning is a new form of the
work; the application of knowledge acquired in school is made at the working place,
and the professional activity is more and more dependent of the intensive-
knowledge. The learning becomes a process inseparable from the adults’ activity in
the process of work. Similarly the children need new educational instruments and
environments that will help them to educate their desire to learn and create. Lifelong
learning is more than “adult education”; it covers and unifies all phases: intuitive
learner (home), scholastic learner (school and university), and skilled domain
worker (workplace) [4].
In the last decade several basic principles of the learning theory have been re-
evaluated, having in view the new facilities offered by the information and
communication technologies evolution, as well as by some lack of success of the
existing e-learning systems. More and more critics sustain that the simple use of ICT
as support of the existing learning practices is insufficient; old frameworks, such as
instructionism, fixed curriculum, memorization, de-contextualized learning etc., are
not changed by technology itself [4].
New computational environment are necessary in order to support new frameworks
for education such as lifelong learning, integration of working and learning, learning
on demand, authentic problems, self-directed learning, information contextualized to
the task at hand, (intrinsic) motivation and collaborative learning. The fulfilment of
the individual needs of each user (expressed explicitly or implicitly) - learning
personalization, educational content re-usability on large scale - content reuse,
assurance of the communication between e-learning systems as well as with other
human resources management systems - interoperability, are the main objectives of
the researches in this domain.
These objectives are the foundation of a complex project SinPers - Innovative
System for Personalized and User-centred Learning with Application to Project
Management. The project is currently developed by: the National Institute for
Research and Development in Informatics (ICI Bucharest), the Academy of
Economic Studies from Bucharest and the Project Management Association from
Romania [5].
The project aims to design and implement an e-learning environment based on
pedagogical concepts and advanced web technologies. It will be able to assist the
building and execution of the whole cycle of a course (from initiation and objectives
definition to the assessment of the results), personalized for each learner, in the
project management field. The choice of „project management” domain for the
application of the new paradigm of learning personalization was not accidental. The
diversity of the experience and knowledge background of the potential users as well
as of the interest, preferences and organizational contexts of those, justify this
option. The survey, carried out in the system requirements specification phase,
confirms entirely this option [7].
Although the system will be first customized for project management (as course
content), the adopted learning model, the content developing methodology
structured on learning objects and the developed software tools will be easily
reusable, for other domains and learning contexts.

2. Personalization - an innovative approach in the e-learning systems
The personalization is an advanced stage in the e-learning systems evolution. The
learners have different learning styles, objectives and preferences, fact that leads to
different efficiency and effectiveness from individual to individual of the traditional
e-learning systems.
The study carried out in the frame of the SinPers project underline that the idea of
personalization can be interpreted and implemented in different ways. Some projects
have focused mainly on the learning style concept (e.g. European project 3DE
Design, Development and Delivery - Electronic Environment for Educational
Multimedia), others on competencies (e.g. European project ADAPT - Adaptive
Design Approach for Personalised Training - Interactive Tools) or on user-context
(e.g. European project LIP Learning in Process) [4].
From the same study results that the learning personalization needs new solutions
for a multitude of aspects, as:
     • adapting to the profile, goals and/or user context / organizational context,
     • knowledge formalization,
     • learner competences or learning objectives definition,
     • learner knowledge acquired diagnosis,
     • adequate feedback generation.
In the frame of SinPers project the personalization problem was solved by adopting
innovative solutions in three main domains: (1) teaching-learning process modelling,
(2) learner modelling and (3) digital content modelling (figure 1).
A useful support in this                           Teaching-learning
approach was the use of IMS                        process modelling
standards that offers a
conceptual frame for all these
three aspects [1]. This choice                      Personalization
was founded on a global                  Learner                        Content
evaluation of existing e-               modelling                     modelling
learning standards [6].
                                       Figure 1. SinPers - personalization pillars

2.1. Teaching-learning process modelling
This activity implied the evaluation of several alternative scenarios and the synthesis
of a composed scenario [8, 9] that includes three main phases (figure 2):
• pre-assessment and the
   personalization of the unit
   of learning based on the
   learner objectives and
• unit of learning execution
• post-assessment, course

                                         Figure 2. SinPers - composed scenario
In order to set up the teaching-learning process specifications we adopted the models
(conceptual, informational and behavioural) recommended by the IMS Learning
Design Information Model standard elaborated by the IMS Global Learning
Consortium. These allow the application and innovation of a diversity of
pedagogical methods as well as the reuse of the educational resources [1].
The conceptual model of a unit of learning (e.g. course, module, lesson) includes:
roles, resources, learning activities (executed by the learner), support activities
(executed by the trainer) and methods. An essential element was the definition of the
method, properties and conditions, on which are based the personalization
mechanisms as well as the control of the process execution.
According to the IMS model, the personalization for each learner is made in several
     • activities tree personalization - through definition of the plays, acts, activity
         structures and role-parts,
    •    environment tree personalization - similar with the activities tree,
    •    educational content personalization (selected learning objects),
    •    personalization according to general user preferences.

2.2. Learner Modelling
The use the learner modelling especially to set up his profile and to direct the
learning process according to this profile is one of the most actual trends [2].
This implies the construction of a qualitative representation that accounts the learner
behaviour at least for two aspects:
     1. existing background knowledge (collected interactively through a direct
         dialogue with the learner, by self-evaluation, or by a pre-assessment test),
     2. knowledge acquired during the learning process.
A third aspect - considered by some advanced e-learning system - is the learner
behaviour during the teaching-learning process.
Generally, the definition of the learner model consists in answers at the following
     • “who” is modelled - identification data, evolution / learner history,
     • “what”- goals, plans, attitudes, capabilities, knowledge, benefits of the
     • “how” - how is acquired and maintained the model,
     • “why” - how are used the data from the model (learning assistance,
         feedback providing, learner behaviour interpretation etc.).
Obviously this model is approximate, partial, qualitative and cannot grasp all the
aspects regarding the learner behaviour. The model setting up can be made
analytically, synthetically or by a combination of these approaches [2].
In SinPers the learner model is built and maintained progressively as the teaching-
learning process steps are covered, through several methods: interactive dialogue,
self-evaluation, guided evaluation controlled by the system or the trainer (sets of
tests followed by grading the obtained results in the pre-assessment phase,
comparison of the tests results with the knowledge set specific to a given
competence or to a given objective followed by the training).
From informational point of view this model is compatible with the IMS Learner
Information Packaging Information Model Specification and contains [1]:
     • general data about the learner (identification, affiliation, studies,
         qualification, competences, course goals, accessibility); these properties
         have and invariant character during the unit of learning execution and can
         be updated only after the unit of learning completion,
     • portfolio data regarding the teaching-learning activities planed for a unit of
         learning, their development stage, durations / deadlines, results,
          intermediary and final grades; these data have a dynamic character and are
          updated under the system and/or trainer control.
The data regarding the affiliation (main activity domain), competences, course goal
and accessibility are essential elements for the personalization of the course and
activities tree.

2.3. Content modelling
The course content and the sequence of the learning activities / assessment of the
acquired knowledge will be presented to the learner, according to its preferences:
     • the entire course,
     • a module of the course,
     • the knowledge remedial about a sub-domain,
     • acquiring of a specified competence,
     • project manager assistance during the development of a project.
In order to meet these options, the SinPers learning materials are structured as a
collection of distinct learning objects (LO). Their reuse in different contexts and
(re)sequencing in different learning paths required the adoption and definition of
two essential aspects:
     • domain ontology (the concepts structure and the relations between them),
     • metadata which describe the learning objects properties.
Subsecquently, the knowledge is represented on different levels of abstractization,
on the lower level are the LOs, defined as entities which may be used, reused or
reffered in the learning process specified before. These are logical conteiners which
represent resources deliverable throught web, like lessons (HTML pages), a
simulation (Java applet), a test (HTML pages with evaluation forms) or any other
object provided through web having learning as goal.
Metadata are a collection of attributes about the objects from the last level, which
are describing the object type (text, slide, simulation, questionnaire etc.), the
requiered educational level (highschool, university etc.), language, interactivity level
and others as well.
The third level of abstractization (ontology) is used for the specification of the
domain concepts and the relations between these. A concept of the domain can be
represented by one or more LOs (having different attributes).
The main relations between concept are: Is_ part_ of and Required_by which
indicate the hierarchical relations between concept and the constrains regarding the
mandatory learning order of the concepts; at these is added optionally the relation
Suggested_Order. The link between the concepts and the learning objects is
explicitated by the relation Explained_by.
3. Conclusions
The lifelong learning becomes an essential pillar of the progress toward the
knowledge society, sustained by the ICT and e-learning systems evolution. The great
diversity of the users of these systems imposes new approaches both from
pedagogical and technological aspects. The fulfilment of the individual requirements
of each user (expressed explicitly or implicitly) - learning personalization, reuse on
large scale of the educational content - content reuse, assurance of the e-learning
systems inter-communication as well as with other human resources management
systems - interoperability, are the main objectives of the present researches. These
objectives are aimed also by the excellence research project SinPers - Innovative
System for Personalized and User-centred Learning with Application to Project
Management (
In order to achieve the main objective of the SinPers project - learning
personalization - it was necessary the research and adopt innovative solutions in
three basic domains: teaching-learning process modelling (e.g. personalization
scenarios, activities flows, events, conditions, properties), learner modelling (e.g.
general and portfolio data, static and dynamic data collection), content modelling
(e.g. domain ontology, learning objects, metadata).

1. IMS Global Learning Consortium, web site at []
2. Wilson, R., Villa, R., (2002) Survey on Methods and Standards for Student Modelling, University of
   Strathclyde Glasgow, UK, IST-2001-33358, Sept., []
3. Milosevic, D., Brkovic, M., Bjekic, D., (2006), Designing Lesson Content in Adaptive Learning
   Environments, International Journal of Emerging Technologies in Learning, ISSN: 1863-0383, vol. 1 (2)
4. SinPers Consortium, (2005) Evaluarea sistemelor actuale, tendinţelor standardelor, Research Report,
   WP1, ICI
5. Trandafir, I., Galatescu, A., Borozan, A-M, (2005), Innovative System for Personalized and User-centered
   Learning with Application to Project Management, Informatica Economică, no. 4 (36), ISSN: 1453-1305
6. Ghilic-Micu, B., Roşca, I.G., Stoica, M., Uscatu, C., Mircea, M., Silvestru, C., (2005), e-learning Standards,
   Informatica Economică no. 4 (36), ISSN: 1453-1305
7. Trandafir, I., Borozan, A-M., (2006), Project Management - Survey Regarding the Experience and
   Training Requirements, Informatica Economică no. 2 (38), ISSN: 1453-1305
8. Trandafir, I., Borozan, A-M., (2006), Proiectarea procesului de învăţare-instruire centrat pe utilizator -
   studiu de caz, Proceedings of the National Conference on Human-Computer Interaction ROCHI 2006,
   Bucharest, 21-22 September, Informatica Economică (supplement) vol. X, ISSN: 1453-1305
9. Trandafir, I., Borozan, A-M., (2007), eLearning Design based on Personalization Requirements,
   International Technology, Education and Development Conference, 7-9 March, Valencia, INTED 2007
   Proceedings CD, ISBN: 978-84-611-4517-1

Shared By: