Learner Models: A Systematic Literature Research in Norms
and Standards
Felix Böck
, Dieter Landes
and Yvonne Sedelmaier
Department of Electrical Engineering and Informatics, Coburg University of Applied Sciences and Arts,
96450 Coburg, Germany
Keywords: Learner Model, Digital Twin, Learner Modeling, Systematic Literature Re view, Systematic Survey,
Systematic Review, Higher Education, Computer Science, Norms, Standards.
Abstract: Learners in higher education tend to become an increasingly heterogeneous group. Paying proper attention to
individual differences is a challenge that may be leveraged by ind ividualized automated recommendations of
learning elements. This presupposes some knowledge of the learners’ profil es which can be captured in so-
called learner models. Yet, so far, there is no comprehensive overview of existing standards and their
contribution related to learner models. This paper presents the resul ts of a systematic literature research
devoted to norms and standards in the area of learner models. As it turns out, 16 norms or standards have
some relationship to learner models, 3 of them present their versions of a learner model. None of the standards
and norms offers a comprehensive learner model, but in their entirety these models provide hints on reasonable
contents and structure of learner models.
Learners in higher education tend to become an
increasingly heterogeneous group. Differences
between learners exist in terms of, e.g., individual
levels of knowledge and competences, varying
learning styles, or individual preferences for (digital)
media. Instructors face severe difficulties in paying
proper attention to these individual differences in
physical classes since individual coaching does not
scale to larger groups of learners.
A potential solution for this dilemma lies in
supplementary offerings of learning materials that are
targeted to the individual needs of a learner in a
specific situation. Yet, such supplementary offerings
face some challenges since learners often do not
know the best learning activity for them in a specific
situation. This implies that mechanisms are needed to
recommend meaningful next activities and materials.
Consequently, such recommendations presuppose
knowledge of properties of individual learners that
would determine the usefulness of learning material
in a specific situation. However, the required
knowledge about the learner is not available,
especially when initialising new (recommendation)
systems (called cold start problem), so that the
possibility of using learning analytics is very limited.
Information systems often have the so-called cold
start problem at the beginning (Bobadilla et al., 2012)
due to a general lack of users within the system
(Schafer et al., 2007; Schein et al., 2002), the learner
is new to the system and no data is yet available (Park
& Chu, 2009; Park & Tuzhilin, 2008) or the learning
element has been newly created and has not yet been
used (Du Boucher-Ryan & Bridge, 2006; Rashid et
al., 2002) i.e. there is no or too little data available
to be able to generate meaningful recommendations
from it. In addition to the behavioural data of the
individual, we believe that the structural
characteristics such as age, previous educational
path, etc. are also important in order to be able to
create a digital image of the learner that is as
comprehensive as possible (Bodily et al., 2018). This
type of knowledge is usually captured in a so-called
learner model or digital twin (Furini et al., 2022;
Hlioui et al., 2016).
Böck, F., Landes, D. and Sedelmaier, Y.
Learner Models: A Systematic Literature Research in Norms and Standards.
DOI: 10.5220/0012556100003693
Paper published under CC license (CC BY-NC-ND 4.0)
In Proceedings of the 16th International Conference on Computer Supported Education (CSEDU 2024) - Volume 2, pages 187-196
ISBN: 978-989-758-697-2; ISSN: 2184-5026
Proceedings Copyright © 2024 by SCITEPRESS Science and Technology Publications, Lda.
Learner models have been discussed for quite
some time, yet without consensus on which contents
should be captured. This paper presents the results of
a systematic analysis of standards and norms that deal
with aspects of learner models in e-learning systems.
With such an analysis, we want to identify
requirements for structure and contents of learner
As it turns out, 16 standards and norms tend to be
relevant. Yet, there is no single source that specifies
the contents of a learner model in a comprehensive
fashion since some of the sources are too specific,
while others cover a variety of aspects in addition to
those related to learner models.
In the following, the paper will clarify important
terms and discuss related work, before section 3
details the employed research approach. Section 4
presents the results of the systematic literature
research. Section 5 concludes the paper and gives an
outlook to future work. A major effort will consist in
combining the results of the analysis of standards and
norms with results from a systematic literature
analysis of scientific papers on learner models that we
conduct in parallel to the work reported here.
Findings of the latter analysis will be presented in a
separate paper.
2.1 Definition of Terms
There is no widely accepted definition of the term
Learner Model”. Rather, discussions of this term are
often controversial. Still, a consistent definition of the
term is necessary as a basis for further work. As a first
step, the two components of the term “Learner
Model” will be considered separately before we
return to the term in its entirety.
A model describes some aspects of a real system
or subject relevant for the respective purpose as a
conscious abstraction of its real counterpart. The most
important characteristics of models are:
Models do not describe the respective system
or subject completely, but from a certain point
of view while neglecting others;
several models can exist for a system or subject
in parallel;
models exist at different levels of abstraction,
ranging from a high-level view to a detailed
all relevant properties of the original must be
adequately and completely mapped to
properties of the model (validation required).
In our field of research, a model serves as a digital
representation of students (learners), with the aim of
providing individual learning support in a digital
“Two things are crucial for an adaptive system to
work: the existence of a means to adapt the task and
the ability to detect the need for adaptation” (Johnson
& Taatgen, 2005, p. 430). This ability to adapt is
generally represented by so-called user models.
Interpretations of the term User Model (UM) differ
widely in the literature (Kay et al., 2022) and depend
on the intended use, scope, domain and the way to
collect information about the user (Hlioui et al.,
2016). In general, a “user model is a representation of
static and dynamic information about an individual
that is utilized throughout the whole interaction
process aiming to trigger a number of adaptation and
personalization effects. […] [This user model] entails
all the information which is considered important in
order to adapt and personalize the user interface
(content and navigation) and functionalities to the
unique characteristics of a user. […] Depending on
the domain and goals of the system, user models can
include different kinds of characteristics about the
users (e.g., interests, preferences, traits, etc.) or data
with respect to their overall context of use (e.g.,
environment, time, interaction device type, etc.)”
(Germanakos & Belk, 2016, p. 79).
Our research focusses on higher education where
the general user model takes a specific shape as
learner model. A Learner Model (LM) thus contains
the individual characteristics and interaction data of a
single learner, represents them explicitly in a
machine-internal representation, and includes many
different aspects, which ultimately depend on the
purpose of the application. The terms Student Model
or Digital Twin, which often appear in the literature,
are synonyms of Learner Model in our setting, as all
our learners are usually university students. The
process of creating such a learner model is called
learner modelling (Khenissi et al., 2015; Piao &
Breslin, 2018).
2.2 Research Question & Field
Even though learner models are relevant for different
disciplines, we explicitly focus here on computer
science (technical implementation) and pedagogy
(teaching methodology) and deliberately ignore other
important sub-disciplines, such as the humanities for
the time being. In doing so, we look at relevant norms
and standards from different points of view from the
perspective of computer science and pedagogy with
the aim of answering the question: “How can
CSEDU 2024 - 16th International Conference on Computer Supported Education
Learner Models look like and what are their
structure and components?”. The selection of the
models to be considered is guided by their (technical)
feasibility, even if they are currently only described
theoretically. Purely theoretical thought experiments
without practical relevance and implementation
possibilities will not be shortlisted.
2.3 Related Work
A widely recognized, well-founded learner model is
still lacking, even though learner models are
becoming increasingly important with the growing
number of e-learning systems and the possibilities of
computer-assisted learning. There is some work on
defining learner models within different projects
(Hlioui et al., 2016; Stiubiener et al., 2010) or for
different purposes (Bodily et al., 2018; Bull & Kay,
2007). A systematic review of scientific literature that
is conducted in parallel to the work reported here will
provide a comprehensive overview of existing learner
models. To the best of our knowledge, however, there
is no systematic analysis of learner models based on
standards yet.
With the increase in digital teaching and learning
opportunities, the desire for uniform standards in the
field of e-learning is growing. The same applies to
user models in digital ecosystems. To get an overview
of related work on the topic of learner models, a
systematic literature review (SLR) according to
(Brereton et al., 2007; Kitchenham & Charters, 2007)
is useful. The process of a systematic literature review
warrants a balanced and objective summary and
overview of the topic of the research question, which
is (theoretically) reproducible and transparent at any
time. Well-defined requirements form the basis for
the five-stage review process: (manual) search,
plausibility check, selection based on filter, check
references of each relevant item and total result. The
planning phase of the systematic literature review
consists of the following general steps.
3.1 Relevant Research Sources
Apparently, there is no single point of contact for
norms and standards, but a wide range of different
providers, from government agencies to consortia of
various commercial companies. Seven popular
electronic database sources [RS1 RS6] were
selected as the most relevant for the area of standards
in e-learning for adaptive teaching and learning. An
important criterion in the selection of data sources
was to obtain published works standards, norms,
drafts or even technical reports from the searched
area that fulfill basic quality criteria (e.g. peer-
reviewed) and are accessible to the public (to a certain
extent). If there are results that cannot be assigned to
any of the above-mentioned publishers, these are
summarized under various and shown separately in
the results. The aim was to identify and evaluate
existing standards in the area of learner models. For
this purpose, current standards, but also discontinued
ones and drafts were used for selection.
Table 1: Overview of Relevant Sources.
World Standards Cooperation (WSC)
International Organization for Standardization
International Electrotechnical Commission (IEC)
International Telecommunication Union (ITU)
IEEE Standards Association (IEEE SA)
1EdTech Consortium
World Wide Web Consortium (W3C)
Internet Engineering Task Force (IETF)
Organization for the Advancement of Structured
Information Standards (OASIS)
This table of providers is supplemented with a
loose collection of other standardised specifications,
for example from ministries/governments or other
organisations that have written guidelines on this
topic. These are grouped under the term “various”.
The spectrum of standards publishers ranges from
international standards publishers to professional
associations and international industry consortia.
As an alliance, the World Standards Cooperation
(WSC) is a globally active association of the
voluntary, consensus-based system of
standardisation. Of the three member organisations,
the two international standardisation organisations
International Organization for Standardization (ISO)
and International Electrotechnical Commission
(IEC) are relevant, which jointly develop
Learner Models: A Systematic Literature Research in Norms and Standards
international standards in the fields of electrics and
electronics. Of the most important professional
organisations in this case, the Association for
Computing Machinery (ACM) and the Institute of
Electrical and Electronics Engineers (IEEE), the
latter develops global consensus-based standards
with the IEEE Standards Association (IEEE SA),
including in the area of learning technology and
information technology. Numerous international
industry partners are also joining forces to form larger
consortia in order to better represent their interests.
During the research phase, four consortia emerged as
an important opportunity. The 1EdTech Consortium
(formerly the IMS Global Learning Consortium) aims
to develop open standards for the e-learning sector. In
recent years, numerous IMS specifications have
become global de facto standards in the field of e-
learning, the issues surrounding networked "Internet
technologies", such as the WWW, are highly relevant.
This is why the two consortia World Wide Web
Consortium (W3C) and the Internet Engineering Task
Force (IETF) are an important starting point in the
search for relevant standards in this area. Less present
organisations, such as the Organisation for the
Advancement of Structured Information Standards
(OASIS), are also represented in this area and should
also be considered.
3.2 Search Strategy
In order to obtain all the desired results and thus to be
able to create as comprehensive an overview of the
domain as possible, it is important in the first step to
create a list of synonyms, abbreviations and
alternative spellings for the selected search terms.
Synonym dictionaries, spelling dictionaries and,
lastly, feedback from our experts were used for this
step. Table 2 shows the search terms used, including
various spellings and abbreviations. This list does not
claim to be complete, but should cover a large part of
the search terms. In the second step, the different
variations are meaningfully linked with each other
using Boolean operators.
The organisations that strive for standards were
searched with a slightly modified search strategy. For
this purpose, the search query above was simplified
and divided into several sequential search queries. If
a search was not possible or not feasible in a
meaningful way, all available standards were
manually sifted through including reading and
assessing the summary in interaction with the title.
Subsequently, the importance for the topics of
eLearning in connection with Learner Models was
checked manually via referenced standards. In
addition, all standards in the eLearning field were
searched by hand so that no important references
were left unnoticed.
Table 2: Search Query Components.
Search Term
"Learner Model"
"Student Model"
"Pupil Model"
"digital Twin"
"educational Data Model"
"User Model"
<Synonym> AND
<Synonym> AND
"Learner Modeling"
"Learner Modelling"
The search terms used for searching the data
sources described above linked with boolean
operators (query) is:
("Learner Model" OR "Student Model"
OR <Synonym>) AND <Alternative>
In addition, a plausibility check was conducted in
the next stage. For this purpose, external sources were
searched for possible results of the publishers, which
were removed by the publisher due to date of
publishing or validity or other reasons and no longer
made available to the public. The aim was to obtain
as comprehensive a view as possible of approaches
that had already been discarded. The results found are
then assigned to the respective publishers.
3.3 Selection Criteria and Filters
To limit the aggregation process, various selection
criteria were defined in order to be able to refine the
existing results in a more targeted manner. The
criteria and various filters are listed and briefly
explained below. Basic filters were
Language of Choice
Only publications in German or English were selected
from the above database resources.
Quality Criteria
Only publications that were subject to quality control
prior to publication were selected. This criterion is
particularly important for drafts and preliminary
standards, as these do not necessarily have to be
subject to standardised quality guidelines. This means
that at least a small group of people worked together
and reviewed each other and additionally a well-
known organization or an association of well-known
CSEDU 2024 - 16th International Conference on Computer Supported Education
commercial companies is behind a standardization
and its concept.
The standard scientific criteria such as title,
keywords, conclusion cannot be transferred 1:1 for
the evaluation of standards and norms. After the
manual search, a manual plausibility check was
carried out. The results found were then filtered
according to
o Abstract
The abstract should already indicate if the
content is appropriate for the research question.
o Research Field/ Discipline
The discipline of the result must be in the field
of e-learning (computer science and teaching
o Full-text
After this pre-selection, the remaining standards
were read fully and evaluated.
After successful filtering, the relevant results were
analysed for their respective references and links. The
relevant results from the references were added to the
result set and ran through the same process described
above. In the case of dubious decisions by the
individual filters, it was assumed that the current filter
was fulfilled, as this “critical” work is then sorted out
by other subsequent filters at a later point in time. The
title was not used as a filter criterion, as experience
had shown that it provided little information about the
content of the relevant standard. The filters are run
sequentially in the order just described. Preferably,
these criteria should be reviewed by multiple
researchers to weed out irrelevant papers. The filters
just described can also be deliberately overridden
individually in very few exceptional cases, for
example if a model found is referenced by numerous
papers (thus constituting seminal work). Such
exceptions are marked separately at first mention.
Furthermore, after the full text analysis, all references
of each relevant article and their authors are checked
to see if there are other important publications outside
the search scope that should definitely be included in
the results.
3.4 Selection of Extracted Data
The data to be extracted from the analysed
publications depend primarily on the respective
research question. The extracted data i.e. the
characteristics of the models - should be as objective
as possible, as these serve as a basis for comparison
between the different models and their features. At
the outset, there should be an awareness that
standards cannot be easily compared. Therefore, the
evaluation should be a purely qualitative
benchmarking of existing standards rather than a
direct comparison between them. Table 3 shows the
most important general attributes of standards. These
can be divided into metadata such as name,
publication date or current status and content data,
in this case the standard’s purpose or its central ideas.
As an option specific attributes may be recorded in
addition to the general attributes.
Table 3: Overview of extracted data categories.
Example values
for attributes
General Attributes
Name of the Standard
Standard for
Publication Date
Last Update
Types of Standard
Specifications (TS)
Computer Science
Revision process
Active Standard
Superseded Standards
Linked Standards
Central ideas / Target
- to standardise a
learner model that is
as generalised as
- basis for the
specialisation of own
4.1 Evaluation of Results
Figure 1 shows the individual (interim) results after
applying the methodology from chapter 3.
From a total of 869 (manually) searched possible
results, only 16 relevant results remain after the
process of systematic literature research. The
continuum of the scope of the standards and norms to
be analysed in detail ranges from around 30 pages of
technical reports to several hundred pages of
international standards. These relevant results are
compared and discussed in the following chapter.
Learner Models: A Systematic Literature Research in Norms and Standards
Figure 1: Overview of the (interim) results of SLR.
CSEDU 2024 - 16th International Conference on Computer Supported Education
4.2 Overview of Results
Before the remaining standards are compared with
each other, they are briefly presented individually
ISO/ IEC 19479:2019 (ISO/IEC 19479, 2019)
Model for recording and exchanging attested learning
achievement information in a formal learning
environment to express the level, content and type of
qualification. In addition, it defines refinements to the
learner mobility achievement award (LMAI) model
for representing the digital diploma supplement,
which is defined in terms of a conceptual model and
a domain model.
ISO/ IEC 19788-9:2015 (ISO/IEC 19788-9, 2015)
Specification for metadata elements and their
attributes for the description of learning resources.
Providing a standards-based approach to the
identification and specification of the metadata
elements required to describe a learning resource.
ISO/ IEC TR 20748-1:2016 (ISO/IEC TR 20748-1,
Specifies a reference model that identifies the diverse
IT system requirements of learning analytics
interoperability. The reference model identifies
relevant terminology, user requirements (use cases),
workflow and a reference architecture for learning
analytics, assessments, accessibility preferences
and data flow and data exchange.
ISO/ IEC TS 29140:2021 (ISO/IEC 29140, 2021)
Provides a learner information model specific to
mobile learning to enable learning, education and
training environments to reflect the specific needs of
mobile participants. The use of a learner information
model for mobile technology in learning, education
and training (mobile learning) is also addressed.
W3C Working Group Note: Making Content
Usable for People with Cognitive and Learning
Disabilities (W3C Working Group, 2021)
Planning, creation and process of accessible
applications usable by people with cognitive and
learning disabilities.
IEEE P1484.2 (IEEE P1484.2, 1997)
Specify the syntax and semantics of a 'Learner
Model', which will characterize a learner and his or
her knowledge/abilities. These elements to be
represented in multiple levels of granularity, from a
coarse overview, down to the smallest conceivable
sub-element and allow also different views of the
Learner Model (learner, teacher, parent, school,
employer, etc.) and substantially address issues of
privacy and security. The Learner Model will provide
more personalized and effective instruction.
IEEE P9274.4.1 (IEEE P9274.4.1, 2022)
Describes the technical implementation of xAPI.
IEEE 9274.1.1-2023 (IEEE 9274.1.1, 2023)
Standardizes the data model format and
communication protocol for learning experience data
allowing vendors to build interoperable solutions and
to take advantage of many products that support the
IEEE P2997 (IEEE P2997, 2021)
Defines the Enterprise Learner Record (ELR) data
model for the various data objects. The ELR data
model preserves data ownership and integrity by
providing indications to where raw learner data is
stored and by providing the ability to track: - the
learner's path through different organizations - the
variety of learning experiences - demonstrated
competencies - conferred credentials - employment
history. Additionally, it defines the transfer methods
and application programming interface (API) for
communicating learner records between services that
adhere to the specification.
1EdTech Learner Information Package
Specification (1EdTech, 2001)
Addresses the interoperability of internet-based
Learner Information systems with other systems that
support the Internet learning environment.
1EdTech Student Learning Data Model (1EdTech,
Provides a complete view of a digital (1EdTech)
ecosystem interconnected with real-time data. This
includes the following sub-areas: User and
Organization, Enrollment and Attendance, Pathways to
Competency, Instructional Resources, Assignment and
Assessment, Learning Activities, Learner Record.
OASIS Specification for JSON Abstract Data
Notation (JADN) (OASIS JADN, 2021)
A formal description technique for expressing the
information needs of communicating applications,
and rules for generating data structures to satisfy
those needs.
Common Education Data Standards (Common
Education Data Standards, 2022)
Data standards and a shared vocabulary for
education data.
Dynamics 365 Education Accelerator (Microsoft
Education, 2023a)
A proprietary and commercial technical
realisation of an education data model component.
Learner Models: A Systematic Literature Research in Norms and Standards
Open Education Analytics (OEA) (Microsoft
Education, 2023b)
An open source-based reference architecture to
develop modern data intelligence capabilities.
Ed-Fi Unifying Data Model (UDM) (Ed-Fi, 2023)
An educational data tool suite (unifying data model,
data exchange framework, application framework,
and sample dashboard source code) that enables vital
academic information on K-12 students to be
consolidated from the different data systems of school
districts so that educators can start addressing the
individual needs of each student from day one, and
can measure progress and refine action plans
throughout the school year. Elements are aligned to
4.3 Discussion of Results
The remaining 16 publications were compared with the
aim of forming a basis for a learner model to be
developed. To this end, various aspects of the learner
model were examined in more detail. Basically, the
standards already differ in terms of the purpose and the
reason for their existence. These range from exchange
options for learner data (system requirements for
interoperability) (ISO/IEC 19479, 2019) (ISO/IEC TR
20748-1, 2016) and their implementation (IEEE
P9274.4.1, 2022), through the specification of learning
resources (ISO/IEC 19788-9, 2015), for example for
people with cognitive and learning disabilities (W3C
Working Group, 2021), to a complete comprehensive
digital ecosystem in which the learner is part of it
(1EdTech, 2020) (Ed-Fi, 2023). The target groups of
the various standards are also broadly diversified and
range from individual learners (ISO/IEC 29140, 2021)
(1EdTech, 2001) to entire groups involved in the
teaching and learning process, for example learners,
teachers, parents and also the administrative
educational institution (IEEE P2997, 2021) (Common
Education Data Standards, 2022). The type and origin
of the standard also differs from, for example, industry
standards (Microsoft Education, 2023a) (Microsoft
Education, 2023b) to active standards from consortia
(IEEE 9274.1.1, 2023) (OASIS JADN, 2021) or their
predevelopment: Drafts (IEEE P1484.2, 1997).
Finally, of the remaining standards, three can be
highlighted that are explicit learner models: (IEEE
P1484.2, 1997) (1EdTech, 2001) and (ISO/IEC
29140, 2021). (ISO/IEC 29140, 2021) describes a
mobile learner model and its specific attributes for
learning, such as device, connectivity, or location.
The (1EdTech, 2001) deals with the
interoperability of Internet-based information systems
for learners with other systems. For this purpose, the
learner is specified and stored in a learner information
server and made available for other applications. The
learner specification is based on a data model that
describes the characteristics of a learner that are
necessary for recording and managing the learner's
progress, goals, achievements and learning experience.
The draft (IEEE P1484.2, 1997) attempted to
specify the syntax and semantics of a learner model
that characterizes a learner and their knowledge/skills
and was first drafted in 1997 with the aim of
centralizing Public and Private Information for
Learners (therefore also known as PAPI Learner).
The learner information is divided into six categories:
PAPI Learning Staff (demographic data)
PAPI Learner Relations (relationships with
other learners)
PAPI Learner Safety (enrollment information)
PAPI Leaner Performance (future goals)
PAPI Learner Preference (preferences)
PAPI Leaner Portfolio (previous experience)
The early date of first publication is a possible
indicator that could point to this, but not a sufficient
condition for the conclusion that the topic of learner
models was already very important at the end of the
1990s. In the early 2000s, the draft was transferred
SC36) in collaboration with IMS (now 1EdTech), but
soon disappeared from the scene. The research could
not uncover any obvious reason why work on the
draft was discontinued.
Research into possible norms and standards in the
area of learner models has shown that there are very
few approaches in this field and that these are either
specialized (ISO/IEC 29140, 2021) or very complex
and more than just learner models (1EdTech, 2001)
or have not yet been pursued further (IEEE P1484.2,
1997). However, these can form a very good basis for
developing your own learner models, such as
(Common Education Data Standards, 2022) or (Ed-
Fi, 2023) and can be extended according to one’s own
For reasons of brevity, not all details can be
presented here. However, in order to ensure the
reproducibility of the SLR and the transparency of the
methodology, all information on the SLR is made
available to the public online (Böck, 2023).
Heterogeneity of learners in higher education is
continuously increasing due to, e.g., individual levels
CSEDU 2024 - 16th International Conference on Computer Supported Education
of knowledge and competences, varying learning
styles, or individual preferences for (digital) media.
Offering learning elements with an optimal fit to the
learner’s specific needs presupposes detailed
knowledge of the learner’s characteristics, captured in
a learner model. A systematic literature research on
existing norms and standards in the area of learner
models revealed 16 relevant publications, 3 of which
present their version of a learner model. Still, these
models are not comprehensive or overloaded with
additional content that does not contribute to
characterizing learners. Nevertheless, these models
provide some indications what should be included in
a learner model and how such a model might be
In parallel to the work reported here, we work on
a systematic literature research and analysis of
scientific publications on learner models. Next steps
will include matching the results of both analyses,
standards on the one hand and scientific literature on
the other, in order to derive requirements on
appropriate contents of such models.
This work is funded by the German Ministry of
Education and Research (Bundesministerium für
Bildung und Forschung) under grant 16DHBKI090
as part of the VoLL-KI project.
1EdTech. (2001). Learner Information Package
Specification (Version 1.0.1). 1EdTech Consortium
(IMS). https://site.imsglobal.org/standards/sldm
1EdTech. (2020). Global Student Learning Data Model
Interconnected Data Powers Better Teaching and
Learning. 1EdTech Consortium. https://site.imsglobal.
Bobadilla, J., Ortega, F., Hernando, A., & Bernal, J. (2012).
A collaborative filtering approach to mitigate the new
user cold start problem. Knowledge-Based Systems, 26,
Böck, F. (2023). All research-relevant attachments from
scientific publications for reproducibility and
transparency [dataset]. GitHub.com.
Bodily, R., Kay, J., Aleven, V., Jivet, I., Davis, D., Xhakaj,
F., & Verbert, K. (2018). Open learner models and
learning analytics dashboards: A systematic review.
Proceedings of the 8th International Conference on
Learning Analytics and Knowledge, 4150.
Brereton, P., Kitchenham, B. A., Budgen, D., Turner, M.,
& Khalil, M. (2007). Lessons from applying the
systematic literature review process within the software
engineering domain. Journal of Systems and Software,
80(4), 571583.
Bull, S., & Kay, J. (2007). Student Models that Invite the
Learner In: The SMILI:() Open Learner Modelling
Framework. International Journal of Artificial
Intelligence in Education, 17(2), 89120.
Common Education Data Standards. (2022). CEDS Data
Model (Version 11) [Draft].
Du Boucher-Ryan, P., & Bridge, D. (2006). Collaborative
Recommending using Formal Concept Analysis.
Knowledge-Based Systems, 19(5), 309315.
Ed-Fi. (2023). Unifying Data Model (UDM) (Ed-Fi Data
Standard for Suite 3 v5.0-pre2). https://techdocs.ed-
Furini, M., Gaggi, O., Mirri, S., Montagero, M., Pelle, E.,
Poggi, F., & Prandi, C. (2022). Digital Twins and
Artificial Intelligence as Pillars of Personalized
Learning Models. Communications of the ACM, 65(4),
Germanakos, P., & Belk, M. (2016). User Modeling. In P.
Germanakos & M. Belk, Human-Centred Web
Adaptation and Personalization (pp. 79102). Springer
International Publishing.
Hlioui, F., Alioui, N., & Gargouri, F. (2016). A survey on
learner models in adaptive E-learning systems. 2016
IEEE/ACS 13th International Conference of Computer
Systems and Applications (AICCSA), 17.
IEEE 9274.1.1. (2023). IEEE Standard for Learning
TechnologyJavaScript Object Notation (JSON) Data
Model Format and Representational State Transfer
(RESTful) Web Service for Learner Experience Data
Tracking and Access. IEEE SA.
IEEE P1484.2. (1997). Standard for Information
TechnologyLearning SystemsLearner Model
(Draft 8). IEEE SA.
IEEE P2997. (2021). Standard for Enterprise Learner
Record. IEEE SA.
IEEE P9274.4.1. (2022). Recommended Practice for the
Implementation of the Experience Application
Programming Interface (API) 1.0.3. IEEE SA.
ISO/IEC 19479. (2019). Information technology for
learning, education, and trainingLearner mobility
achievement information (LMAI). ISO/IEC.
ISO/IEC 19788-9. (2015). Information technology
Learning, education and trainingMetadata for
learning resourcesPart 9: Data elements for persons.
ISO/IEC 29140. (2021). Information technology for
learning, education and trainingNomadicity and
mobile technologies (Version 1). ISO/IEC.
ISO/IEC TR 20748-1. (2016). Information technology for
learning, education and trainingLearning analytics
interoperabilityPart 1: Reference model. ISO/IEC.
Johnson, A., & Taatgen, N. (2005). User Modeling. In R.
W. Proctor & K.-P. L. Vu (Eds.), Handbook of human
Learner Models: A Systematic Literature Research in Norms and Standards
factors in Web design (pp. 424439). Lawrence
Erlbaum Associates.
Kay, J., Bartimote, K., Kitto, K., Kummerfeld, B., Liu, D.,
& Reimann, P. (2022). Enhancing learning by Open
Learner Model (OLM) driven data design. Computers
and Education: Artificial Intelligence, 3, 100069.
Khenissi, M. A., Essalmi, F., Jemni, M., & Kinshuk.
(2015). Learner Modeling Using Educational Games: A
Review of the Literature. Smart Learning
Environments, 2(1), 6.
Kitchenham, B. A., & Charters, S. (2007). Guidelines for
performing Systematic Literature Reviews in Software
Engineering (EBSE 2007-001). Keele University.
Microsoft Education. (2023a). Dynamics 365 Education
Accelerator (v5.0.3.1). https://learn.microsoft.com/en-
Microsoft Education. (2023b). Open Education Analytics
(OEA) (v0.1). https://openeducationanalytics.org
OASIS JADN. (2021). Specification for JSON Abstract
Data Notation (JADN) Version 1.0 (Committee
Specification 01). Organization for the Advancement of
Structured Information Standards. OASIS.
Park, S.-T., & Chu, W. (2009). Pairwise preference
regression for cold-start recommendation. Proceedings
of the Third ACM Conference on Recommender
Systems, 2128.
Park, S.-T., Pennock, D., Madani, O., Good, N., &
DeCoste, D. (2006). Naïve filterbots for robust cold-
start recommendations. Proceedings of the 12th ACM
SIGKDD International Conference on Knowledge
Discovery and Data Mining, 699705.
Park, Y.-J., & Tuzhilin, A. (2008). The long tail of
recommender systems and how to leverage it.
Proceedings of the 2008 ACM Conference on
Recommender Systems, 1118.
Piao, G., & Breslin, J. G. (2018). Inferring user interests in
microblogging social networks: A survey. User
Modeling and User-Adapted Interaction, 28(3), 277
Rashid, A. M., Albert, I., Cosley, D., Lam, S. K., McNee,
S. M., Konstan, J. A., & Riedl, J. (2002). Getting to
know you: Learning new user preferences in
recommender systems. Proceedings of the 7th
International Conference on Intelligent User
Interfaces, 127134.
Schafer, J. B., Frankowski, D., Herlocker, J., & Sen, S.
(2007). Collaborative Filtering Recommender Systems.
In P. Brusilovsky, A. Kobsa, & W. Nejdl (Eds.), The
Adaptive Web (Vol. 4321, pp. 291324). Springer
Berlin Heidelberg.
Schein, A. I., Popescul, A., Ungar, L. H., & Pennock, D. M.
(2002). Methods and metrics for cold-start
recommendations. Proceedings of the 25th Annual
International ACM SIGIR Conference on Research and
Development in Information Retrieval, 253260.
Stiubiener, I., Vicente, W., & Rosatelli, M. (2010). An
Approach in Personalisation and Privacy in E-Learning
Systems. In M. Buzzi (Ed.), E-learning. InTech.
W3C Working Group. (2021). Making Content Usable for
People with Cognitive and Learning Disabilities.
World Wide Web Consortium (W3C).
CSEDU 2024 - 16th International Conference on Computer Supported Education