Towards a Self-Regulated Learning in a Lifelong Learning
Perspective
Nour El Mawas
1
, Jean-Marie Gilliot
1
, Serge Garlatti
1
, Patricia Serrano Alvarado
2
, Hala Skaf-Molli
2
,
Jérôme Eneau
3
, Genevieve Lameul
3
, Jacques-Francois Marchandise
3
and Hugues Pentecouteau
3
1
IMT Atlantique, UMR CNRS 6285 LabSTICC, Brest, France
2
Université de Nantes, LS2N, Nantes, France
3
Université Rennes 2, CREAD, Rennes, France
Keywords: Lifelong Learning, Self-Regulated Learning, Professional Development, Open Learner Models, Semantic
Web.
Abstract: The professional development presents many difficulties related to speed of change and the explosion of
knowledge that requires people to learn at many intervals throughout their lives. This study proposes a
combined Self-Regulated Learning Process, functional and technical architectures in a Lifelong Learning
perspective. The Self-Regulated Learning is carried out using Semantic Open Learner Models. We illustrate
our process through some services examples. This work is dedicated to the Lifelong Learning active
community and more specifically to researchers in Technology Enhanced Learning, pedagogical engineers,
and learners who meets difficulties in integrating multidisciplinary expertise, technology and know-how
throughout their life.
1 INTRODUCTION
Nowadays, a person will have many different jobs
during his/her life. Lifelong Learning is becoming a
central asset, beginning during initial training at
university, pursuing during the whole career with
many different jobs. Learning is also Life wide, as it
occurs in multiple, formal and informal contexts:
school, home, work, etc. Lifelong and Lifewide
Learning are key elements for the prosperity,
especially in a knowledge society.
Adult education research acknowledges that most
of Lifelong Learning outcomes are acquired apart
from formal learning (whether they come from school
or university). Non formal, informal or incidental
learning represent the vast majority of adult learning.
In other words, self-learning methods (self-directed
learning, self-documentation, meetings with fellows
or relatives, etc.) constitute the main majority of
learning resources (Tremblay 2003). Whether they
are planned or not, related to a concrete goal or
acquired on the fly due to exchange or pure
coincidence, these resources link the knowledge to
experience (from life, work, etc.). This is tacit,
implicit, informal knowledge, which allows everyone
to build autonomously his own experience and his
own learning path.
Current learning traces and learning recognition
are very scarce and formal. The capitalization passes
via different tools, devices and methods: portfolios,
training personal account, recognition of personal and
professional experience, etc. A challenge today is to
organize those resources in order to scaffold self-
learning process and to explore new learning models
that can help to capture implicit, informal knowledge
in order to capitalize the different training
experiences and work throughout the life.
The motivation behind this work is the speed of
change and the explosion of knowledge that requires
people to learn at many intervals throughout their
lives. For this reason, schools and universities are no
longer providing a package of knowledge and skills
to serve a person for life. Learners need to ensure their
professional development in a Lifelong Learning
perspective.
The main objective of this paper is to define
combined process and functional / technical
architectures for personal Lifelong Learning. The
process and architectures will allow learners i) to
Mawas, N., Gilliot, J-M., Garlatti, S., Alvarado, P., Skaf-Molli, H., Eneau, J., Lameul, G., Marchandise, J-F. and Pentecouteau, H.
Towards a Self-Regulated Learning in a Lifelong Learning Perspective.
DOI: 10.5220/0006387506610670
In Proceedings of the 9th International Conference on Computer Supported Education (CSEDU 2017) - Volume 1, pages 661-670
ISBN: 978-989-758-239-4
Copyright © 2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
661

formalize their experience on tacit knowledge, ii) to
capitalize their knowledge and competences, and iii)
to foster their collaborative interactions and social
knowledge building.
Section 2 proposes the theoretical background of
the study. Section 3 presents several existing
approaches for professional development and
Lifelong Learning dimensions. Section 4 details our
scientific positioning including our Self-Regulated
Learning Process, Semantic Open Learner Models,
and Technical infrastructure. Section 5 defines our
evaluation perspectives in the research project named
Sedela. Section 6 summarizes the conclusion of this
paper and presents its perspectives.
2 THEORETICAL
BACKGROUND
The basic idea behind the term "Lifelong Learning" is
that learning can and should occur through each
person's lifetime (Knapper and Cropley, 2002). The
basic premise of Lifelong Learning (Sharples, 2000)
is that it is not feasible to equip learners at school,
college or university with all the knowledge and skills
they need to prosper throughout their lifetimes.
Therefore, people will need continually to enhance
their knowledge and skills, in order to address
immediate problems and to participate in a process of
continuing vocational and professional development.
The new educational imperative is to empower people
to manage their own learning in a variety of contexts
throughout their lifetimes (Bentley, 1998). The
European Lifelong Learning Initiative defines
Lifelong Learning as “a continuously supportive
process which stimulates and empowers individuals
to acquire all the knowledge, values, skills and
understanding they will require throughout their
lifetimes and to apply them with confidence,
creativity and enjoyment, in all roles circumstances,
and environments” (Watson, 2003).
Professional development (Day 1999) consists of
all natural learning experiences and those conscious
and planned activities which are intended to be of
direct or indirect benefit to the individual, group or
school and which contribute, through these, to the
quality of education. It is the process by which, alone
and with others, learners review, renew and extend
their commitment as actors of their learning; and by
which they acquire and develop critically the
knowledge, skills and emotional intelligence essential
to good professional thinking, planning and practice
through each phase of their lives.
Professional development is a crucial aspect of
Lifelong Learning. Indeed, the best way to support,
develop, and cultivate an attitude of Lifelong
Learning is through a professional development
focusing on learners needs identified by them, their
institutions, and their communities. In other words,
professional development is a way to improve the
quality of learning and develop a culture for Lifelong
Learning.
3 RELATED WORKS
In this section, we consider existing approaches and
infrastructures to manage professional development
and what are important dimensions of Lifelong
Learning that we must take into account in order to
insure this professional development.
3.1 Existent Approaches for
Professional Development
The state of the art shows us mainly existing
approaches and infrastructures to manage
professional development and related learning. We
mention Learning Analytics (LA), Personal
Knowledge Management (PKM), Personal Learning
Environment (PLE), e-portfolio (e-P), Personal
Knowledge Network (PKN), and finally the LinkedIn
platform (LI).
Learning Analytics (Gasevic et al. 2015) is the
Big Data approach for learning. Learning analytics is
the measurement, collection, analysis and reporting
of data about learners and their contexts, for purposes
of understanding and optimizing learning and the
environments in which it occurs. Many Universities
or educational services, such as MOOC platforms, are
collecting data to conduct analysis. This is however
bounded to an institution or a commercial platform.
Data collection and analysis are conducted according
to internal services with no connection to students’
needs. Moreover data cannot be long term stored and
must be used for predefined analysis for ethical
reasons.
PKM and PLE (Chatti 2007) relate to a collection
of processes that a person uses to gather, classify,
store, search, retrieve, and share knowledge in his or
her daily activities and the way in which these
processes support activities. The former is more
dedicated to work, while the latter may embed
additional services related to learning. In both cases,
it is based on the idea that persons need to be
responsible for their own growth and learning, and
LLL 2017 - Special Session on Lifelong Learning
662

push a bottom-up approach. However, those
approaches capitalize on knowledge production
rather on personal development.
PKN (Chatti et al., 2012) considers the Learning
as a Network (LaaN) based on connectivism,
complexity theory, and double-loop learning. LaaN
starts from the learner and views learning as the
continuous creation of PKN. For each learner, a PKN
is unique adaptive repertoire of tacit and explicit
knowledge nodes (people and information) and one’s
theories-in-use (norms for individual performance,
strategies for achieving values, and assumptions that
bind strategies and values together).
A portfolio is a meaningful documentation of a
learning path, either for assessment or for formative
purposes (Ravet 2007). E-Portfolios (e-P) are one of
those tools that have been appeared in education since
Internet usage becomes more widespread. Compared
with paper based portfolios, they also have the added
values in terms of keeping records, connecting ideas,
relating information, and publication (Barrett 2006).
However, in existing implementations, recording is
manual. It relates to reflexive process, not to current
learning support, and no data access granting for other
purpose than collaborative reflexion is provided.
LinkedIn (LI) is a business oriented social
networking service. It provides a powerful cloud-
based CV service, and professional networking
services, where every data is public. LinkedIn
exploits user data collected to provide valuable
information. It already proposes higher education
curriculum ranking, according to job wishes, based on
alumni analysis. It totally control algorithms used,
analysis derived, and how information is provided,
without direct feedback to users.
Our interest is to link the professional
development to the Lifelong Learning. That is why
we present in next sections the Lifelong Learning
dimensions and the existing approaches regarding
these dimensions.
3.2 Lifelong Learning Dimensions
Researchers discuss important dimensions in the
Lifelong Learning (Narciss et al., 2007) (Sloep et al.
2011). These dimensions are: capitalization of
learning experiences including work and long term
learning, learning recognition, learning goals
management, personal learning management, and
social learning.
In a long term perspective, the capitalization of
learning experiences should be provided. It has to be
able to manipulate different data: learning traces,
learning evidence, learning confidence, professional
outcomes, and recommendations. Learners will need
to organize them, and evaluate achievements. This
self-managed database should be organized to
support Self-Regulated Learning Process (SRLP),
according to relevant Learner Models.
Out of this database, we pay a special attention to
learning recognition: diplomas, certificates,
recommendations as they constitute external support
of SRLP. They acknowledge achievements and
constitute certified evidence.
Learning goals management is a key for SRLP. It
is a very personal decision that has its roots in a social
environment providing examples, discussions and
opportunities.
To reach these goals, learners need to plan, to
conduct and to regulate their learning process. All of
this is a personal learning management that can be
instrumented, i.e., modeled.
Collaboration is essential to support learning;
hence our last dimension is social learning. It includes
the ability to share and to interact with others, and to
contribute to emerging knowledge.
Following these dimensions, we will discuss in
the next section the positioning of each existing
approach for professional development regarding
these dimensions.
3.3 Discussion
In this section, we examine existing approaches
detailed in section 3.1 and we analyze if they take into
account the Lifelong Learning dimensions (Table 1).
Table 1: Comparison between existing professional
development approaches and LLL dimensions.
dim1 dim2 dim3 dim4 dim5
LA 1
course
- - 1
course
-
PKM - - - - x
PLE - - x x x
PKN - - x x x
e-P x x - -
LI - x - - x
For the sake of clarity, dim1 refers to the
capitalization of learning experiences, dim2 to
learning recognition, dim3 to learning goals
management, dim4 to personal learning management,
and dim5 to Collaboration. LA allows the
capitalization of learning experience and the personal
learning management for only one course. PKM
ensures collaboration through sharing knowledge.
PLE and PKN take into account the learning goals
management, the personal learning management, and
Towards a Self-Regulated Learning in a Lifelong Learning Perspective
663

collaboration. E-P provides the capitalization of
learning experiences and learning recognition. LI
guarantees the learning recognition and collaboration
between professional (through peers’
recommendation).
Across the table 1, we found that no existing
approach for professional development meets our
Lifelong Learning dimensions.
The needed approach should be a support to the
capitalization of learning experiences. It must also
insures the learning recognition and promote learning
goals management, personal learning management,
and collaboration. This led us to think deeply about a
new approach that insures the professional
development in a Lifelong Learning perspective.
4 OUR APPROACH
In a Lifelong Learning perspective, learner
empowerment relates to the ability for a person to be
able to define his/her own learning path and act on
his/her environment, including peer learning. It is a
prerequisite for autonomy, to deal with many
different jobs and corresponding learning
requirements during his/her whole career. Our
approach aims at designing, developing,
experimenting and evaluating an improved model of
Self-Regulated Learning Process, supported by
Semantic Open Learner Models and an experimental
infrastructure, in a Lifelong Learning perspective.
Firstly, the concept of Self-Regulated Learning
Process is introduced. Secondly, the Open learner
models and its benefits to support Lifelong Learning
are presented. Thirdly, how the SOLM is built.
Fourthly, the relationship between SOLM and
learning Services is discussed. Finally, the technical
infrastructure is detailed.
4.1 Self-Regulated Learning Process
Our main goal is to experiment learning methods
developing Self-Regulated Learning Process in a
lifelong learner autonomy perspective and to provide
new opportunities for professional and self-
development based on the concept of portfolio,
recommendation, and quantitative / qualitative data
collection that can be “aggregated” in an open learner
model.
The Self-Regulated Learning Process refers to the
learner himself. Since the early research on the
autonomous learner in the 1980s, work has shown the
close links between autonomy, reflexivity and
metacognition (Tremblay, 2003), and more especially
between self-directed learning, self-determination
and self-regulation (Schunk and Zimmerman, 1998;
Cosnefroy, 2001). These models show that
motivation and personal project (individual
components of learning) are essential but not
sufficient elements: the psychosocial dimensions of
learning (through collaboration, trust, evaluation) are
decisive for building effective learning environments.
Indeed, if the learning process is personal, it can
nevertheless be taught, with adapted learning
methods. Although the former work of Schön (1983)
on "reflective practitioner" has been widely used for
vocational training, it has only gradually irrigated
Lifelong Learning issues (in work/study programs,
higher education, etc.). The same holds true taking
into account the role of personal experience in
professionalization process. More recently, the
question of reflexivity emerged regarding the
evolution of higher education within the impact of
competences in the curricula reforms and the
emergence of new devices and methods to take into
account the learning experience (Rege-Colet and
Berthiaume 2015).
Nowadays, the theoretical models of learners’
empowerment are numerous and pursue different
ends, sometimes competing (Eneau, 2012 2016): self-
regulation of learning (for learning management),
self-directed learning and self-development (for
Lifelong Learning) and even awareness and critique
(for transformative learning). These different models
have been widely discussed, especially for the
assumptions on which they are based and for the
instrumentation they can generate.
On a methodological level, the complexity and
subtlety of learning situations, where a set of different
dimensions (cognitive, affective, biographical) are
mobilized, require to articulate the joint analyses of
researchers and learners together, and even the
involvement of teachers and pedagogical advisers.
This means participatory action research (PAR) or
design-based research, within individual and
collective inquiries to analyze the activity (through
self-confrontation, explicitness, focus groups, etc.).
This crossed methodology should then respond to
different objectives of analysis, understanding and
transformation of learning methods and help to
develop some new and more integrative tools
(Lameul and Loisy 2014).
The main consideration is that learner
empowerment cannot be solely based on the control
or reinforcement of technical skills or abilities. On the
contrary, methods and tools must be thought in terms
of capacitation (Falzon 2013; Oudet 2012), i.e., (i) the
support they provided to each learner, individually,
LLL 2017 - Special Session on Lifelong Learning
664

for self-orientation, production of informed choices,
of personal and professionals projects, etc.; (ii) the
possibility they offer to anyone to become aware of
his/her strengths and weaknesses, for improvement or
enhancement, in a perspective of automatization; (iii)
the purpose of these policies of capacitation in terms
of equity, so as to guarantee to everyone equal
opportunities, in terms of capabilities (Eneau &
Simonian 2015).
To some extent, tools thought in term of
capacitation have to fulfill the different Lifelong
Learning dimensions proposed in the paragraph 3.2.
This requires making the learning process more
explicit by means of adequate learner models
providing relevant self-information. The next section
shows how Open Learner Models can support the
learning process and the corresponding Lifelong
Learning dimensions.
4.2 Semantic Open Learner Models
A learner model refers to the model constructed from
observation of interaction between a learner and a
Technology Enhanced Learning system. “The learner
model is a model of the knowledge, difficulties and
misconceptions of the individual. As a student learns
the target material, the data in the learner model about
their understanding is updated to reflect their current
beliefs” (Bull, 2004).
An Open Learner Model (OLM) makes a
machines’ representation of the learner available as
an important means to support learning (Bull & Kay
2010). It can be viewed or accessed by the learner, or
by other stakeholders (e.g. teachers, peers, parents,
etc.). In the Learner Model community, important
studies have been developed about learner control,
understandability, availability of various sources,
visualizations and their impact on learning (Bull &
Kay 2016).
Indeed, there are a variety of ways in which an
open learner model might be useful to the learner and
support the Lifelong Learning process: (i)
Capitalization of learning experiences and Learning
recognition: the open learner model can be built and
updated from a large variety of data sources (user
interactions, learning analytics, badges, evidences
from exercises or QCM, diplomas, certificates,
badges, endorsements like in LinkedIn,
recommendations, etc.; (ii) Learning Goals
management: Promoting metacognitive activities
(reflection, planning, self-monitoring); (iii) Personal
learning management: Allowing the learner to take
greater control and responsibility over their learning,
encouraging learner independence; facilitating
navigation to materials, exercises, problems or tasks,
etc., where links are available from the learner model;
increasing learner trust in an adaptive educational
environment by showing the system’s inferences
about their knowledge; (iv) Collaboration: Prompting
or supporting collaborative interactions amongst
groups of students, Facilitating interaction between
learners and peers, teachers and parents; Etc.
In others words, an Open Learner Model is a
necessary machine representation of the learner
knowledge and skills and their progression to support
Lifelong Learning. In a recent publication (Bull and
Kay 2016), the SMILI framework for interfaces to
learning data in open learner models has been
reported. This paper has also studied the different
works of the entire OLM community. The community
is focused on the design of OLM environments to
access and modify short term learner models. Various
usages of such open learner models are proposed in
the literature, but no generic infrastructure enabling
personal data access, long term storage and data
transfer (Gilliot et al. 2016). This exhibits that
Lifelong Learning perspective is still an emerging
question in the Learner Model community.
To support Self-Regulated Learning Process, a
learner model has to be partially shared and
exchanged with different stakeholders, compared to
the resources metadata to support adaptation,
compared to other learner models to support
collaboration, etc. In other words it is necessary to
ensure interoperability among resource metadata,
learner models and other models supporting learning
processes. An Open Learner Model based on
Ontologies and the semantic web principles will
enable us to provide interoperability at knowledge
level among collaborative services, Open Learner
Models and distributed data sources and trusted
services.
Thus, we will design an innovative semantic
OLMs (SOLM) to give more expressive power to
facilitate self-regulation and systematic development
of collaborative services tailored for self-directed
learning.
Semantic Open Learner Models will be developed
to support Self-Regulated Learning by making
informal or incidental learning resources more
explicit. This will be achieved by capturing,
managing, sharing, etc., personal learning data from
various heterogeneous sources with semantic
enhancement. Semantic models will enable long term
management and a certain level of trust for
collaborative services. It enables us to explicit
reasoning and learner model usages. In our context,
personal learning data are data owned and controlled
Towards a Self-Regulated Learning in a Lifelong Learning Perspective
665

by people themselves to deal with learning services
addressing Lifelong Learning dimensions.
4.3 Building the SOLM
The SOLM building will be based on an iterative
process. The process will be as follows: i) First of all,
a state of the art in the domain of learner models
(OLM, Lifelong Learner modeling ...), portfolios,
competences (IMS, 2008), and standards like IEEE
PAPI (Oubahssi and Grandbastien, 2006), IMS LIP
(Kalz, 2007) will be established. According to that
study, a first version of the SOLM and its
corresponding ontologies will be designed and/or
reused; ii) According to a use case, its
experimentations and the corresponding data
gathering, iterative SOLM improvements will be
defined according to test-field feedbacks till having a
“stable version”; iii) The different SOLM versions
will be aggregated. The main idea is to get a
consolidated version of SOLM that ensure reusability
and interoperability among a set of uses cases.
4.4 SOLM and Learning Services
To exemplify how we address the Lifelong Learning
dimensions linked to an open learner model, we
define some micro scenarios that are needed to
develop students “meta-competences” (Tremblay,
2003). These scenarios are learner oriented, as it is
depicted thanks to “my” possessive determiner. The
reflexive learning process will be developed by
combining those micro-scenarios:
1. “My knowledge” depicts the need to aggregate
different sources of data: whether learning traces,
learning evidence, professional outcomes,
recommendations, that will constitute personal
resources for reflexive process. This learning service
should combine manual entries and automatic
recording to provide the Open Learner Model
properties. A learner can scrutinize, control and
manage his open learner model to provide a certain
level of trust.
2. “My CV”: the learner is able to select a view of
“his knowledge” to publish some relevant
information, a “CV”, to cooperate with others (such
as a peer in a project), to apply for a new job or a new
training. This learning service enables the user to
manage its open learner model to provide a CV, a
specific portfolio
3. “My learn in progress” refers to the ability to
organize learning, i.e., to have learning goals,
construct a learning todo list with tasks done, access
to current working documents.
4. “My collaborations” are the person or the
services a learner wishes to collaborate with.
Collaboration is a win-win process based on resource
sharing, data integration and trust. The open learner
model sharing owner will manage grant access and
usage policies to control data usage and to develop
trustful collaborations.
5. “My Future Course” is a combination of
recommendations by the community and a wish list.
Recommendations could be for example for a
software engineer trainings of new technologies
(MongoDB, ElasticSearch…) identified as relevant in
his community. The wish list is a translation of
continuing development. This services is based on the
open learner features that must be shared, exchanged,
compared, etc. By granting access of his open learner
model and to the community learner models, a learner
can manage its future course.
6. “Alumni feedback” is an example of external
service that could provide fruitful collaboration. By
granting access of his open learner model to an
alumni community, a learner can get back relevant
information, such as possible occupations and
corresponding learning paths.
According to these micro-scenarios, table 2
presents the comparison between our proposed
services and LLL dimensions. For the sake of clarity,
S1 refers to “My knowledge”, S2 to “My CV”, S3 to
“My learn-in-Progress”, S4 to “My Collaborations”,
S5 to “My Future course”, and S6 to “Alumni
Feedback”.
Table 2: Comparison between our proposed services and
LLL dimensions.
In the next section, we define our technical
infrastructure in order to provide the SRLP carried
out using Semantic Open Learner Models and some
services examples.
4.5 Technical Infrastructure
Our infrastructure is a Semantic PIMS (SPIMS)
(Abiteboul 2015) in the cloud. In a previous
publication (Gilliot et al., 2016), we have presented a
proof-of-concept prototype based on a Personal cloud
LLL 2017 - Special Session on Lifelong Learning
666

infrastructure and standard interface implementation
to collect data.
Figure 1 shows the prototype architecture (Gilliot
et al., 2016). Components of the personal cloud are
highlighted in green. Learning components
(including services) are highlighted in violet. In our
architecture, e-Portfolio (1) is seen as an example of
Learner Model. It enables data access according to
lifelong personal goals. Implementing such Learner
Model in a Personal Cloud provides personal data
storage (2), enabling full data access to the learner
and full duration control as well. Data are collected in
two different ways: external learning achievements
may be collected thanks a data transfer mechanism
(3) from external servers, whether institutional or
commercial, or learning traces thanks a learning
streaming flow (4). The proxy mechanism (5)
provides a basic mechanism to grant access
selectively.
Figure 1: PIMS.
In this context, we developed two data transfer
connectors. The first data transfer connector retrieve
Open Badges, where the user may synchronize his
personal learning achievement database with existing
backpack. As validation of badges is maintained by
external (institutional) servers, the user is only able to
classify which ones are relevant for what purpose in
his e-Portfolio. Other digital diplomas can be
retrieved in a similar way. The second data transfer
connector retrieve commercial e-Portfolios, the
commercial e-Portfolio service provides a specific
API enabling download of existing learner
certifications. This service can be extended in the case
of LinkedIn.
Once the data transfer connectors are
implemented, we need to aggregate data from various
learning sources, this must be achieved through
specific API, based on linked data to enable higher
semantic information level, or data streams. Those
data are collected in data stores, providing access to
various services see (Gilliot et al., 2016) like
reflection, visualization, adaptive learning…. New
standards have emerged, called xAPI that provides
data streams based on statements (ex “I did this”) to
depict activities, and on Learning Record Stores
(LRS) to provide data access. Those standards are
widely adopted in the open learning environments
(Santos et al., 2015). In our context, statements are
duplicated in the learner personal cloud and the
external LRS, enabling data collection for personal
(4.1) and institutional (4.2) record storage at the same
time. This gives the opportunity to fulfill institutional
analytics needs, and give direct access to the user as
well. Our architecture also enables the exchange
between personal and institutional records.
We developed a specific Learning Record Store
compatible with cozy framework and used the xAPI
to enable data aggregation from various contexts. As
it is embedded in cozy context, it ensures user control,
as well as the ability to fine grained control access to
third party services and to other LRS.
Figure 2: Our infrastructure.
This prototype is able to store statements from
various applications proposing a xAPI wrapper. We
used some basic examples, and developed a specific
wrapper we tested on nQuire, which is a personal
inquiry learning system proposed by the Open
University (Mulholland et al., 2012). As a proof of
concept, this wrapper sends activity statements to the
personal LRS of the user and in parallel to an
institutional LRS.
Figure 2 shows our infrastructure. There are two
main roles of the SPIMS. The first role is to collect
learning data via xAPI. These data are provided from
different sources like self-declaration, peer
recommendation, badges, certificates, diplomas,
professional experience, informal learning, and
Towards a Self-Regulated Learning in a Lifelong Learning Perspective
667

Learning Analytics. The second role is to control the
use of the learning data (via proposed services) by the
learner himself. These data must comply with the
semantic models from the SOLM, the domain
models, the activities (SRLP) and data sources model.
Services use also these semantic models in order to
respond to requested queries, to use data for inference
purposes, and to insure Human-Machine interactions.
We have already explained that semantic aspect is
very important in our approach to support Self-
Regulated Learning.
Figure 3 shows the interaction between different
PIMS. Each PIMS controls the access to its own
learning data stored, retrieved and manipulated in
Resource Description Framework (RDF). If PIMS 1
want to access to specific data of PIMS 2, a SPARQL
request is sent from PIMS 1 to services that call PIMS
2 and depending on its authorization the requested
data will be shared or not with PIMS 1. More
generally, it will be necessary to process federated
queries over the distributed RDF data sources to
provide the different learning services (Montoya et al.
2015).
Figure 3: Collaboration and semantic aspect in our
approach.
The proposed architecture has to address the
following features: heterogeneity of data,
interoperability and scalability. Heterogeneity of data
and interoperability can be ensured by a semantic web
approach (Ontologies and Linked Data). It is one of
the main role of the semantic web. Nevertheless, the
level of abstraction vs specialization of the different
ontologies have to be chosen carefully to foster the
reuse of SOLM and its usefulness – more level of
abstraction leads to more reusability, but less
usefulness. It is also necessary to deal with the use of
very different data sources and to limit its impact on
the redesign. In terms of scalability, a difficult
problem will address: a federated query engine at
scale up. Indeed, existing federated query engines
cannot scale for a large number of data sources. We
want to propose a federated query engine that can
scale for a large number of SPIMS.
After having described our approach, we will
proceed to our evaluation perspectives in the next
section.
5 EVALUATION PERSPECTIVES
In the context of our approach evaluation, we have
identified apprenticeship training as key experiment
field, including professional environment and
reflexive approach. Working with students will
enable us to track progress in Self-Regulated
Learning Process, to qualify learning methods and to
test our tool’s based on our infrastructure as an
intertwined experiment. Those experimentations will
be conducted on two different populations: (1)
Apprenticeship students in education science, where
professional development is a central asset in the
curriculum; (2) Engineering students.
Innovative learning methods will be defined in
coordination of professors in charge of professional
development of the two populations. Qualitative
interviews will be conducted with teachers and
learners as well, in order to identify relevant
advancements and potential additional needs in this
new kind of personal environments. Experimental
rounds will be based on a semester period, and we aim
at being operational during two full academic years.
For the tool’s test phase and the analysis of the
practices, two methodologies will be developed.
The first methodology is a qualitative
methodology in two stages:
a. The technique will be the semi-directed
interview (Blanchet and Gotman 2015) in a
comprehensive epistemological framework
(Kaufmann 1996). The goal is here, from an
individual collection of data to create a typology of
the situation and current practices. For follow up
interviews, we‘ll use the principles of the explicitness
(Vermersch 2010);
b. On the other hand, we’ll collect group
interviews (Duchesne and Haegel 2008) with students
in training to estimate the variation of the practices
with regard to the purposes of the training.
The second methodology is about a questionnaire
that will be sent out to collect a second level of
objectification (Martin, 2005), to measure:
a. Various understanding of the tools and
procedures,
b. The students’ satisfaction compared with
the training objectives.
In parallel, additional indicators could be
designed thanks to data collection in experimental
infrastructure. Such indicators will be developed as
trusted collaborative services.
LLL 2017 - Special Session on Lifelong Learning
668

6 CONCLUSIONS
This study addresses the problem of the speed of
change and the explosion of knowledge that requires
people to learn at many intervals throughout their
lives. The main questions of the study are how to
address, what are the approaches allowing the
professional development in a Lifelong Learning
perspective, and how to promote Self-Regulated
Learning. We investigate the problem from its
theoretical background, and we consider existing
approaches for the professional development in order
to see if any existing approach can meet our
requirements. Unfortunately no one can respond to
our needs in terms of capitalization of learning
experiences including work and long term learning,
learning recognition, learning goals management,
personal learning management, and social learning.
To achieve this, Self-Regulated Learning Process is
proposed with functional and technical solutions to
our problem. This solution allows learners to insure
their Self-Regulated Learning, to manage their data
learning and collaborate with peers.
Our perspectives are that our Self-Regulated
Learning Process will take advantage of explicit Open
Learner Models to develop and support self-learning
methods in personal and professional development.
Trusted and long term capitalization will enable
lifelong perspective. Collaborative services will
provide the needed socialization for Lifelong
Learning and organizational knowledge creation.
REFERENCES
Abiteboul S., André B., Kaplan D. (2015). Managing your
digital life with a Personal information management
system. Communications of the ACM, 2015, 58 (5),
pp.32-35.
Barrett, H. C. (2006). Using electronic portfolios for
formative/classroom-based assessment. Classroom
Connect Connected Newsletter, 13(2), pp. 4–6.
Bentley, T. (1998). Learning Beyond the Classroom:
Education for a Changing World. London: Routledge.
Blanchet A. & Gotman A. (2015). L’entretien, Paris,
Armand Colin.
Bull, S. (2004). Supporting learning with open learner
models. Planning, 29(14), 1.
Bull, S., & Kay, J. (2016). SMILI☺: a Framework for
Interfaces to Learning Data in Open Learner Models,
Learning Analytics and Related Fields. International
Journal of Artificial Intelligence in Education, 26(1),
293-331.
Chatti, M. A., Schroeder, U., & Jarke, M. (2012). LaaN:
convergence of knowledge management and
technology-enhanced learning. Learning Technologies,
IEEE Transactions on, 5(2), 177-189.
Chatti, M.A., M. Jarke, and D.Frosch-Wilke, (2007). The
future of e-learning: a shift to knowledge networking
and social software International. Journal of
Knowledge and Learning, 2007. 3(4/5): p. 404-420.
Cosnefroy, L. (2001). L'apprentissage autorégulé, entre
cognition et motivation. Grenoble : PUG.
Day, C. (1999). Developing teachers, the challenge of
Lifelong Learning, Falmer Press, London, Philadelphia.
Duchesne S. & Haegel F., (2008). L’entretien collectif,
Paris, Armand Colin.
Eneau, J. & Simonian, S. (2015). Cultures d’apprentissage
et scénarios pédagogiques : concilier différence et
diversité. Questions de Communication, 29, 49-68.
Falzon, P. (2013). Pour une ergonomie constructive.
Ergonomie constructive, 1-15.
Gašević, D., Dawson, S., & Siemens, G. (2015). Let’s not
forget: Learning analytics are about learning.
TechTrends, 59(1), 64-71.
Gilliot, J.-M., El Mawas N., & Garlatti S. (2016). "Towards
Personal infrastructure to manage long term open
learner models." PALE 2016: Workshop on
Personalization Approaches in Learning
Environments, part of 24th ACM Conference on User
Modeling, Adaptation and Personalisation (UMAP
2016).
Gilliot J.M., Garlatti S., Rebaï I., Pham Nguyen C. (2012).
A Mobile Learning Scenario improvement for HST
Inquiry Based learning. Workshop Emerging Web
Technologies, Facing the Future of Education.
IMS, “IMS Reusable Definition of Competency or
Educational Objective (RDCEO),” http://
www.imsglobal.org/competencies/index.html, 2008.
Kalz, M. “Positioning of Learners in Learning Networks
with Content, Metadata and Ontologies,” Interactive
Learning Environments, vol. 15, no. 2, pp. 191-200,
2007.
Kaufmann J-C (1996). L’entretien compréhensif, Paris,
Nathan.
Knapper, C.K. and Cropley, A. J., (2000). Lifelong
Learning and Higher education, London: Croom Helm.
Lameul, G. & Loisy, C. (2014).
La pédagogie universitaire
à l'heure du numérique. Bruxelles : De Boeck
Supérieur.
Martin O. (2005). L’enquête et ses méthodes, l’analyse de
données quantitatives, Paris, Armand Colin.
Mulholland, P., Anastopoulou, S., Collins, T., Feisst, M.,
Gaved, M., Kerawalla, L., & Wright, M. (2012).
nQuire: technological support for personal inquiry
learning. Learning Technologies, IEEE Transactions
on, 5(2), 157-169.
Narciss, S., Proske, A. and Koerndle, H. (2007). ‘Promoting
Self-Regulated Learning in web-based learning
environments’, in Computers in Human Behavior,
23(3), pp.1126–1144.
Oubahssi, L. and Grandbastien, M. “From Learner
Information Packages to Student Models: Which
Continuum?” Intelligent Tutoring Systems, pp. 288-
297, 2006.
Ravet, S. (2007). Position Paper on ePortfolio (Avaliable
Towards a Self-Regulated Learning in a Lifelong Learning Perspective
669

from:http://www.eifel.org/publications/eportfolio/docu
mentation/positionpaper, accessed: 20.02.2017).
Rege Colet, N. & Berthiaume, D. (Eds) (2015). La
pédagogie de l’enseignement supérieur: repères
théoriques et applications pratiques. Berne: Peter
Lang.
Santos, J. L., Verbert, K., Klerkx, J., Duval, E., Charleer,
S., & Ternier, S. (2015). Tracking Data in Open
Learning Environments. Journal of Universal
Computer Science, 21(7), 976-996.
Schunk, D. H., & Zimmerman, B. J. (Eds.). (1998). Self-
regulated Learning: From Teaching to Self-reflective
Practice. New York- London: The Guilford Press.
Schön, D. A. (1983). The reflective practitioner: how
professionals think in action. New York: Basic Books.
Sharples, M. (2000). The design of personal mobile
technologies for Lifelong Learning (2000). Computers
and Education, 34, pp. 177–193.
Sloep, P., Boon, J., Cornu, B., Klebl, M., Lefrere, P., Naeve,
A., & Tinoca, L. (2011). A European research agenda
for Lifelong Learning. International Journal of
Technology Enhanced Learning, 3(2), 204-228.
Tremblay N. A., (2003). L’autoformation - Pour apprendre
autrement, Presses de l’Université de Montréal,
Montréal.
Vermersch P. (2010). L’entretien d’explicitation, Paris,
ESF.
Watson, L. (2003). Lifelong Learning in Australia,
Canberra, Department of Education, Science and
Training.
LLL 2017 - Special Session on Lifelong Learning
670