Support Learning Design and Analytics with EduP Knowledge Model

Thi Hong Phuc Nguyen

1,2

, Ngoc Tram Nguyen-Huynh

1,2

and Thi My Hang Vu

1,2,*

Faculty of Information Technology, University of Science, Ho Chi Minh City, Vietnam

Vietnam National University, Ho Chi Minh City, Vietnam

Keywords: Knowledge Model, Ontology, Learning Design, Learning Analytics.

Abstract: Learning design (LD) have been a prominent topic in the academic community for many years. It aims at

planning and organizing learning activities and resources to promote learning process and engage students in

achieving learning outcomes. Learning analytics (LA) has matured in the education field and developed a

strong connection with learning design. Learning analytics provides valuable insights to inform learning

design decisions, while learning design serves as a means to turn learning analytics results into actionable

strategies. Their alignment completes the big picture for enhancing teaching and learning. Despite numerous

studies proposing means to support LD/LA and their alignment, both fields still face many challenges due to

the lack of a consolidated framework for reflecting on the various types of knowledge essential for LD/LA.

This paper aims at proposing a comprehensive framework, named EduP (Education-Domain-User-

Pedagogy), that supports LD/LA by leveraging different types of knowledge. The main contributions of the

framework include a knowledge model and an insight engine. The knowledge model helps clarify essential

components for LD/LA and their relationships, while the insight engine addresses how this knowledge is

accessible to teachers in the context of LD/LA. A brief discussion on the implications and future research is

also presented.

1 INTRODUCTION

Our research focuses on the multidisciplinary aspects

of learning design and learning analytics, and their

alignment within a framework that supports them by

leveraging knowledge-based solutions.

Learning Design (LD) has been a prominent

topic in the academic community for many years. LD

focuses on creating and refining learning scenarios,

i.e., which consist of a sequence of learning activities

and resources to engage learners in achieving specific

learning outcomes (Koper & Bennett, 2008).

The creation of these scenarios is based on various

pedagogical strategies (e.g., preparing activities and

resources for problem-based learning differs from

those used in inquiry-based learning). Many studies

have been conducted to assist teachers in creating

effective learning scenarios through various means,

such as editing tools (Celik & Magoulas, 2016; Pozzi

et al., 2020), modeling languages for specifying

learning scenario elements (Botturi & Stubbs, 2008),

or pedagogical design patterns (Eyal & Gil, 2020).

Corresponding author

Learning Analytics (LA) is another area of

interest in educational science, focusing on

collecting, processing, and mining educational data to

generate insights that support educational decision-

making (Hernández-de-Menéndez et al., 2022; C.

Romero & Ventura, 2020). With the advent of

advanced data analytics tools and methods, learning

analytics (LA) has emerged as a powerful tool for

analyzing educational data to enhance learning and

teaching (Mangaroska & Giannakos, 2019).

Many researchers in the analytics field focus on

developing learning analytics dashboards to visualize

learner performance and progression (Susnjak et al.,

2022). Others concentrate on identifying associations

within learning data to discover new insights, such as

predicting dropout rates or identifying at-risk learners

(Ouyang et al., 2023; Ramaswami et al., 2023).

Another area of interest is personalization, which

aims to provide learners with appropriate activities

and resources based on their learning contexts (Chatti

& Muslim, 2019; Romero et al., 2019).

Learning design and analytics have matured in

their respective fields. However, these two topics

Nguyen, T., Nguyen-Huynh, N. and Vu, T.

Support Learning Design and Analytics with EduP Knowledge Model.

DOI: 10.5220/0013019600003838

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 16th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2024) - Volume 3: KMIS, pages 97-107

ISBN: 978-989-758-716-0; ISSN: 2184-3228

have a strong convergence. Learning analytics

provides valuable insights to inform learning design

decisions. On the other hand, learning design serves

as a means to turn learning analytics results into

actionable strategies (Mangaroska & Giannakos,

2019). Together, these two fields offer a

comprehensive approach to improving teaching and

learning activities. As a result, a significant number

of studies focus on the alignment of learning design

and learning analytics (Ahmad et al., 2022; Bakharia

et al., 2016).

Despite the numerous studies on proposing means

to support LD/LA and their alignment, both fields still

faces many challenges. The first challenge involves

elaborating learning scenarios, which requires

teachers to have a strong understanding of both

pedagogical principles and learning domains, as well

as how to integrate them effectively (Schmitz et al.,

2017). A significant number of teachers lack this

knowledge, which limits their ability to design

effective learning scenarios (Lui & Bonner, 2016;

Tatto et al., 2020). Therefore, developing learning

scenarios remains a challenging task for teachers that

requires additional support (Vu & Tchounikine,

2021). The second challenge involves the lack of a

consolidated framework for learning analytics, which

prevents data from being interpreted meaningfully.

This makes it difficult to derive actionable insights

from the data, thereby complicating their effective

application to enhance teaching and learning (Ahmad

et al., 2022).

To address these challenges, this study aims to

propose a comprehensive framework, named EduP

(Education-Domain-User-Pedagogy), that supports

learning design and analytics by leveraging different

types of knowledge, such as pedagogical knowledge

and learning domain knowledge. The framework also

focuses on ensuring that this knowledge is accessible

to teachers in the context of learning design and

learning analytics.

To propose such supportive tools, knowledge

should be clarified and structured efficiently.

Additionally, to provide a consolidated framework,

all stages—from collecting and organizing data to

importing it into the knowledge base, to exploiting

and disseminating the knowledge to teachers—

should be well-defined. Ontologies, which provide

formal representations of domain concepts and serve

as powerful reasoning tools, are essential for

structuring knowledge (Vu et al., 2023). The state-of-

the-art reveals numerous types of ontologies for

modeling learning activities, learning outcomes,

learning domain knowledge, learner profiles, or

generic ontologies that can be applied across various

domains (Rahayu et al., 2022; Wang & Wang, 2021).

In the subsequent sections, the paper presents in

more detail an ontology-based framework for LD/LA.

These sections are organized as follows. Section 2

presents the methodology adopted to target the

objectives of this study. Section 3 introduces related

works by first providing a brief summary of essential

topics on knowledge modeling and exploiting in

education, followed by a review of the state-of-the-art

research in these areas. Section 4 clarifies the first

output of the paper, which is the definition of

essential knowledge types that assist in LD/LA.

Section 5 introduces a knowledge structure to

organize these knowledge types as the second output

of the paper. Subsequently, Section 6 provides a

method/process for discovering this knowledge

through the use of a reasoning engine as the last

output of the paper. Section 7 focuses on the

validation of the propsed framework through some

real-world scenarios in higher education. Finally,

Section 8 concludes by the implications and

limitations of the study and suggests directions for

future research.

2 METHODOLOGY

This section outlines a methodology based on the

Design Science Research (DSR) methodology to

conduct the research presented in this paper (Dresch

et al., 2015). DSR emphasizes the creation of

innovative artifacts to solve specific problems.

DSR’s artifacts can be: constructs providing

fundamental concepts for describing a specific

problem and its solutions; models linking the

constructs in a real-world situation; methods

providing guidelines/processes for solving problems;

and instantiations demonstrating how the theoretical

constructs, models, and methods can be applied in

practice (Peffers et al., 2007).

This research aims at proposing a knowledge-

based framework for enhancing learning design and

analytics (EduP Framework). The artifacts for the

framework are created through the following phases.

Problem Identification. This phase focuses on

identifying the research questions to be addressed for

building EduP framework. Two key questions are

identified: RQ#1: What types of essential knowledge

can support learning design and analytics? And

RQ#2: How can the knowledge be elaborated and

used effectively?

Solution Definition. This phase defines the

objectives of a solution to solve the identified

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

problem, which requires the EduP framework: define

essential knowledge types in LD/LA and determine

effective methods for reasoning and disseminating

this knowledge. To define these objectives, a brief

literature review is conducted to summarize the

current state-of-the art in knowledge modeling and

reasoning for the education sector.

Design and Development. This phase involves

creating EduP artifacts. These artifacts are classified

in constructs, models, methods, and instantiation,

according to DSR methodology (Peffers et al., 2007).

 EduP Constructs and EduP knowledge model

are proposed to address RQ#1. The constructs

define key knowledge components in LD/LA,

while the knowledge model outlines how these

components are related to one another.

 EduP Insight Engine is proposed as a

method/process within the framework to

response to RQ#2. The method defines

multiple modules for representing, elaborating,

and reasoning about knowledge, aiming to

generate insights in LD/LA.

 Two Instantiations are also created to validate

the framework in a subsequent phase. The first

one is an ontology based on EduP knowledge

model. The second one is a web-based reasoner

built upon EduP insight engine. The reasoner

serves as a prototype for reasoning with the

created ontology through a simple interface.

Demonstration and Evaluation. This phase

involves validating the proposed framework in real-

world situations. The ontology and web-based

reasoner developed in the previous phase are used to

address various case studies in the higher education

context.

3 RELATED WORKS

This section provides an overview of current research

on ontology-based solutions for knowledge modeling

and reasoning in education, addressing both the

structural and behavioral aspects of these solutions.

3.1 Structural Aspect

The structural aspect focuses on the types of

knowledge represented in ontologies. An ontology is

a specification that defines concepts within a domain

and their relationships in a structured, formal, and

explicit manner (Gruber, 1993). In education, an

ontology is defined as “a system of primitive

vocabularies/concepts for constructing a tutoring

system” (Mizoguchi et al., 1996). In technology-

enhanced learning, ontologies are considered

effective tools for modeling the learning and teaching

domain due to their formal expressiveness, support

for sharing, and reasoning capabilities.

In General, ontologies can be used to model a

wide variety of information types. The classification

of these ontologies can follow different criteria, such

as their levels of abstraction (domain-independent

and domain-specific ontologies) (Guarino et al.,

2009), their intentions (domain ontologies modeling

a target domain, task ontologies modeling generic

problems and their solutions, and application

ontologies dedicated to activities within a specific

application), (Al-Yahya et al., 2015; Mizoguchi et al.,

1996). From the perspective of smart systems,

ontologies can be classified in five major types:

ontologies for modeling generic concepts across

domains, domain ontologies, user ontologies, context

ontologies, and merged ontologies combining

multiple ontology types to provide a comprehensive

reasoning (Chimalakonda & Nori, 2020; Vu et al.,

2023).

In Education, most ontologies are domain

ontologies, which are used to describe the concepts of

specific learning domains such as mathematics,

physics, and programming (e.g., Iatrellis et al., 2019;

Lalingkar et al., 2014; Ramesh et al., 2016). Other

research also focuses on modeling pedagogical

elements such as curriculum/syllabus, learning

scenarios, learning activities, and outcomes

(Hyunsook & Jeongmin, 2016; Katis et al., 2018;

Reynolds et al., 2023). Learner profiles specifies user

data such as user profiles, interest, needs; which is

typically employed for learning recommendations

and personalization (e.g., Pelap et al., 2023; Romero

et al., 2019). Context ontologies are another type that

emerged with the evolution of smart systems,

supporting retrieving the most relevant knowledge,

according to a specific learning context (Aguilar et

al., 2018; Cabrera et al., 2017; Ouissem et al., 2021;

Perera et al., 2014).

3.2 Behavioral Aspect

The behavioral aspect examines the processes and

methods used to develop and reason with knowledge.

This involves two key components: elaboration and

reasoning.

Elaboration, also known as ontology building,

involves the extraction of data and its integration into

a predefined knowledge model or ontology structure,

as outlined in the structural aspect. which can be done

manually, semi-automatically, and automatically.

Support Learning Design and Analytics with EduP Knowledge Model

Manual approaches are performed by domain

experts who analyze the specific domain, annotate

data, and manually integrate it into an ontology

structure (e.g., Verdú et al., 2017). This method is

costly and prone to errors, especially with large

datasets. However, human interaction and expert

domain analysis can ensure the rationality of the

extracted data, which is essential for complex

domains such as education.

Automatic approaches employ natural language

processing (NLP), data mining, or machine learning

algorithms to extract information from unstructured

data, such as text, and integrate it into an ontology

structure without human intervention (e.g., Aguilar et

al., 2018; Lacasta et al., 2018; Wei & Shao, 2022).

These methods facilitate efficient and cost-effective

knowledge extraction from large datasets. However,

the absence of expert monitoring can lead to the

generation of knowledge that may be unreasonable or

inaccurate within the domain.

Semi-automatic approaches involve both domain

experts and algorithms (e.g., (Cano-Benito et al.,

2021; Chang et al., 2020; Ghazal et al., 2020; T. M.

H. Vu & Tchounikine, 2021)). Several parts of the

process are performed or supervised by experts, while

others are carried out by algorithms. These

approaches benefit from AI techniques to automate

part of the extraction process, while also leveraging

the collaboration with domain experts to minimize the

risk of constructing an incomplete or potentially

incorrect knowledge base

The second key component of the behavioral

aspect is reasoning. The purpose of reasoning is to

derive insights, make inferences, and effectively

utilize the knowledge within ontologies to address

issues and answer questions. This can be done

through various methods such as query languages,

built-in reasoners, and user-defined inference rules

and algorithms.

Ontology query languages and built-in reasoners

are typical solutions for ontology-based reasoning.

The most widely used ontology query language is

SPARQL, which provides a formal syntax for

extracting and manipulating data within ontologies

(e.g., Lacasta et al., 2018; LeClair et al., 2022; Wen

et al., 2022). Another solution involves using built-in

reasoners such as HermiT and Pellet, which are

integrated into ontology editors like Protégé (e.g.,

Andrade et al., 2019). The reasoners enable automatic

deduction and consistency checking. However, these

approaches have a limitation: they require additional

interfaces to be user-friendly, as query languages and

the Protégé interface are too complex for non-

technical users.

Another common solution involves using human-

defined rules to discover knowledge from ontologies

(e.g., Bensassi et al., 2019; Ghazal et al., 2020; L.

Romero et al., 2019) or proposing custom solutions

and algorithms tailored to specific applications and

purposes (e.g., Demaidi et al., 2018). These

approaches can be resource-intensive, requiring

significant input from experts to define rules and

program algorithms, and may not fully leverage the

inherent benefits of ontology support.

3.3 Research Gap Identification

From a brief summary and analysis of related works,

we have identified several research gaps that

highlight the motivation behind our research and

provide a clear direction for how our work can

address the current limitations in the existing

literature. These limitations are outlined as follows.

There is Insufficient Focus on Aligning

Multiple Ontologies to Enhance Teaching and

Learning. Improving LD/LA requires teachers to

have a strong knowledge not only in their specific

teaching domain but also in pedagogical strategies to

create effective learning scenarios. Aligning these

ontologies should be clearly defined and integrated

into a unified framework that can holistically

facilitate their implementation and application in

LD/LA.

There is a lack of user-friendly tools for

utilizing knowledge from the end-user perspective.

Numerous types of ontologies have been proposed,

but we still lack an efficient means to deliver the

knowledge contained within these ontologies to users.

Current methods do not pay enough attention to

bridging the gap between the complex, structured

data within ontologies and the practical, accessible

insights needed by end users. This highlights a

critical need for developing user-friendly interfaces

and tools that can facilitate the efficient extraction

and application of knowledge from ontologies.

There is a Lack of a Unified Framework That

Incorporates a Knowledge-Based Approach to

Promoting LD/LA. Such a framework would

facilitate a more cohesive and systematic application

of ontological principles and could serve as a

blueprint for the future conception and development

of these services. The framework needs to encompass

all phases, from collecting data and integrating it into

knowledge bases to delivering the knowledge

effectively to end users through efficient means.

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

100

4 KNOWLEDGE COMPONENTS

This section introduces EduP constructs in detail to

clarify the key knowledge components involved in

knowledge-based learning design and analytics.

These constructs are organized into a multi-level

structure to facilitate reusability and future extension.

The abstract level adopts the 5W1H model (who,

what, why, when, where, and how) as proposed by

(Jang & Woo, 2012). This model enables general

reasoning to anwser the question such as “who

achieves what in which context (when, where)?”.

Inherited from the abstract level, concepts at

lower levels are tailored specifically to the education

sector. This level encompasses three core knowledge

types in a KB-based LD/LA framework: pedagogical

knowledge (learning goal, learning outcome, learning

level, activity); learning domain knowledge (topic);

users (individual learner, group). Additionally,

contextual knowledge is defined to facilitate

connections among these three knowledge types (see

Figure 1 for the proposed constructs and their

relationships).

5 KNOWLEDGE MODEL

This section defines a generic knowledge structure

that we use to organized the proposed constructs.

Pedagogy-Pedagogy Linking involves

connecting various aspects of teaching and learning

to ensure that educational strategies are aligned with

learning goals (the relations “includes”, “achieves”,

“targets”, “involves”). It encompasses two main

components: first, defining the learning goals, which

are the specific objectives students are expected to

achieve; and second, developing learning activities

that help students meet these goals through

measurable learning outcomes. The connection

between learning outcomes and activities is mediated

by the context in which learning occurs,

Figure 1: Knowledge Model.

Support Learning Design and Analytics with EduP Knowledge Model

101

emphasizing that the design of activities should be

adapted to the specific "when" and "where" of the

learning environment. For instance, the activities

chosen to achieve a particular learning outcome may

vary between in-class sessions and homework

assignments, reflecting the need for context-sensitive

pedagogical approaches.

Domain-Pedagogy Linking involves integrating

pedagogical knowledge with learning domain content

to enhance learning analytics (the relations

“hasLearningGoal”, “hasLearningOutcome”). This

process connects various elements of pedagogy—

such as learning outcomes, learning goals, and

instructional activities—with the content-specific

knowledge relevant to a course. By linking these

pedagogical components with the learning domain

knowledge, educators can better track and analyze

what topics learners have mastered. This alignment

supports the effective measurement of student

progress and achievement, facilitating more accurate

and actionable insights into learning outcomes. It

helps in identifying which specific topics students

have successfully learned and which areas may

require additional focus, thereby enabling targeted

interventions and improved educational strategies.

Domain-Pedagogy-Learner Linking involves

integrating pedagogical and learning domain

knowledge with individual learner profiles to enhance

educational experiences. This approach aligns

learning outcomes, goals, and activities with learner

assessments and enrollment contexts. This integration

also enables sophisticated analytics to track which

topics learners have mastered and at what level of

proficiency. Furthermore, it allows for the creation of

personalized tools that enable learners to monitor

their own performance and progres.

6 INSIGHT ENGINE

This section details the EduP insight engine, proposed

as a method/process within the EduP framework to

address RQ#2: how can knowledge be elaborated and

used effectively? The method defines multiple

modules for representing, elaborating, and reasoning

about knowledge, aiming to generate insights in

Learning Design (LD) and Learning Analytics (LA).

To propose a unified approach that covers the

entire process from handling raw data to delivering

insights and to highlight the transitions of data to

knowledge and knowledge to insights, the process

proposed here relies on the DIKW (Data,

Information, Knowledge, Wisdom) model, which is

hierarchical framework that allows structuring

processes in a systematic way (Rowley, 2007).

Accordingly, EduP Insight Engine composes of

the three main components, as presented in Figure 2.

 Data Module: Organize and preprocess data to

be imported into the ontology.

 Information Module: Create an ontologie

from the data processed in the Data Module.

 Knowledge Module: Reason with the

ontology to generate knowledge for predefined

specific use cases (Wisdom).

This is a semi-automatic process that involves

some actions performed automatically by programs

(denoted by rectangles in light red) and others

requiring user intervention (denoted by rectangles in

light blue). The details are presented below.

Figure 2: Insight Engine.

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

102

6.1 Data Module

The objective of this module is to collect, process, and

transfer data from different sources into data files in

CSV format, based on predefined structures (file

templates). These data files will then be imported into

the proposed ontology structure in the information

module.

Data Source. Three primary data sources are

considered here: syllabus documents used to identify

pedagogical knowledge (learning outcomes, learning

goals, learning activities) and a portion of the learning

domain knowledge (topics to be acquired by

learners); data from LMS used to acquire user history

and interaction; databases from institutions providing

learner assessment information.

Organization. This submodule aims to transfer

data from various sources into CSV files. Some parts

of this submodule require intervention from educators

or teachers, particularly in identifying essential

pedagogical knowledge from textual syllabus.

Processing. This submodule involves

preprocessing the data to prepare it for import through

Python programs. Tasks include handling missing

values, removing duplicates, and merging or splitting

files if necessary. The output is a properly formatted

spreadsheet that can be automatically imported into

the ontologies.

6.2 Information Module

This module is responsible for creating ontologies

(also known as knowledge bases) from the data

processed in the data module. This ensures that the

data is structured in a way that facilitates reasoning in

the knowledge module. Protégé is used as the editor

for constructing these ontologies.

Ontology Loading. This submodule begins by

creating the ontology's abstract structure based on the

EduP knowledge model in Protégé. This structure

includes the main concepts and their relationships and

remains nearly unchanged throughout the ontology's

lifecycle. In subsequent steps, ontology data from

CSV or Excel files will be imported into this abstract

structure.

Rule Definition. These are JSON-based rules that

define the mapping between the content of

CSV/Excel files and the ontology structure. They

specify how each part of the data files can be

identified and mapped into a component of the

ontology structure. Since the data is complex and

large, automatic loading using these predefined rules

is essential.

6.3 Knowledge Module

This module focuses on exploring ontologies to

generate insights and subsequently delivering these

insights to other systems through APIs.

Reasoning Engine. This submodule consists of a

set of SPARQL queries that can be used to reason the

ontology created from the information module. The

query system is defined based on predefined use cases

analyzed and proposed from the perspectives of

educators and teachers.

Business Case Prediction. These predefined use

cases are declared as elements in the wisdom module.

For each specified use case, the required knowledge

is identified, and appropriate queries are invoked.

This module is responsible for mapping the

predefined use cases to suitable knowledge. This task

is currently conducted through collaboration between

educators/teachers and programmers.

Knowledge Deliver. This submodule manages

the interaction between the EduP Insight Engine and

other applications, such as web-based interfaces,

through APIs. It handles receiving requests and

responding to them, enabling the delivery of insights

to educators and teachers. This allows for visualizing

results in an accessible format, facilitating informed

decision-making and enhancing the educational

process.

6.4 Wisdom Module

This module is not a software component. Instead, it

results from requirement analysis from the

perspectives of educators and teachers. The

predefined use cases capture essential knowledge for

common requirements in learning design and

analytics (see Figure 3 for more details).

Figure 3: Wisdow Management Module.

Data Management. This group enables end-users

import, export, and modify ontology data through

user-friendly interfaces.

Knowledge Reasoning. This group supports

teachers and educators in reflecting on various types

Support Learning Design and Analytics with EduP Knowledge Model

103

of knowledge. Each case study outlines potential

outputs based on the input knowledge type. For

instance, given a specific learning outcome for a

course, the related data might include the associated

learning goal, relevant topics, the required learning

level according to Bloom's Taxonomy, and the

learning activities designed to achieve this outcome.

Understanding these types of knowledge is crucial for

assisting teachers and educators in learning design.

However, the conception and development of

learning design tools is beyond the scope of this

research.

Learning Analytics. This group focuses on data

analytics based on knowledge types defined in

ontologies. For example, analyzing the connection

between learning outcomes and learner assessment

results can generate statistics on which learners

achieve or do not achieve the outcomes.

7 INSTANTIATIONS

This section presents two instantiations developed to

validate the EdUP framework. First, the EduP

knowledge structure and process are applied to create

an ontology for a database course, demonstrating the

framework's applicability. Second, the EduP web-

based reasoner is constructed to illustrate how the

ontology can support learning design and learning

analytics through real-world scenarios.

The reasoner

processes the constructed ontologies to generate

insights for teachers through interactive user

interfaces.

As illustrated in Figure 4, the reasoning process

provides results related to specific learning outcomes.

By default, all relationships associated with a

specified learning outcome are presented in a simple

table format, which facilitates easy consultation for

teachers. This user-friendly presentation allows

educators to quickly access and interpret relevant

information, aiding in the evaluation and refinement

of teaching strategies. The interactive nature of the

interface enhances the usability of the insights,

making it easier for teachers to leverage data in their

decision-making processes.

Figure 5 displays the interface for descriptive

analytics, which highlights the number of students

who have completed specific activities within a

course. This straightforward statistical overview

demonstrates how the knowledge extracted from

ontologies can be effectively applied in learning

analytics. By providing clear metrics on student

participation and achievement, the interface

showcases the practical value of integrating

ontological knowledge into educational analysis. This

integration not only aids in tracking student progress

but also illustrates how such knowledge can be used

to derive actionable insights for improving course

design and instructional strategies.

Figure 4: Knowledge Reasoning Interface.

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

104

Figure 5: Learning Analytics Interface.

8 CONCLUSIONS

This section summarizes the contributions of the

paper and offers suggestions for future research

directions.

In terms of contribution, the paper first presents

a comprehensive knowledge model that integrates

various types of knowledge within the context of

LD/LA. This model provides teachers with a holistic

overview of how domain-specific knowledge can be

acquired through various pedagogical strategies. The

second contribution is a method that defines the main

phases and associated components to facilitate

reasoning on knowledge bases. By linking multiple

knowledge types and providing a structured method

for reasoning on knowledge bases, this paper offers

valuable tools for educators and researchers. The case

studies used for validation highlight the potential for

implementing the proposed framework in the future.

In terms of future research, since the framework

is a proof of concept, the knowledge model is

currently simple and needs further development to

meet the requirements of LD/LA. Additionally, some

components of the proposed method can be

automated to reduce costs, leveraging advancements

in AI, for example, automatically identifying learning

topics from syllabus. Finally, more research is needed

to explore how knowledge can be translated into

measurable indicators within learning analytics.

ACKNOWLEDGEMENTS

This research is funded by University of Science,

VNU-HCM under grant number CNTT 2023-09.

REFERENCES

Aguilar, J., Jerez, M., & Rodríguez, T. (2018). CAMeOnto:

Context awareness meta ontology modeling. Applied

Computing and Informatics, 14(2), Article 2.

https://doi.org/10.1016/j.aci.2017.08.001

Ahmad, A., Schneider, J., Griffiths, D., Biedermann, D.,

Schiffner, D., Greller, W., & Drachsler, H. (2022).

Connecting the dots – A literature review on learning

analytics indicators from a learning design perspective.

Journal of Computer Assisted Learning, jcal.12716.

https://doi.org/10.1111/jcal.12716

Al-Yahya, M., George, R., & Alfaries, A. (2015).

Ontologies in E-Learning: Review of the Literature.

International Journal of Software Engineering and Its

Applications, 9(2).

Andrade, L. O. M., Valter, R., Ramos, R., Vidal, V.,

Andrade, D., & Oliveira, M. (2019). LARIISA: An

intelligent platform to help decision makers in the

brazilian health public system. Proceedings of the 25th

Brazillian Symposium on Multimedia and the Web,

501–504. https://doi.org/10.1145/3323503.3362122

Bakharia, A., Corrin, L., De Barba, P., Kennedy, G.,

Gašević, D., Mulder, R., Williams, D., Dawson, S., &

Lockyer, L. (2016). A conceptual framework linking

learning design with learning analytics. Proceedings of

the Sixth International Conference on Learning

Analytics & Knowledge - LAK ’16, 329–338.

https://doi.org/10.1145/2883851.2883944

Support Learning Design and Analytics with EduP Knowledge Model

105

Bensassi, I., Elyusufi, Y., & El Mokhtar, E.-N. (2019).

Smart Connection of User Profiles in a Big Data

Context. Proceedings of the 4th International

Conference on Big Data and Internet of Things, 1–8.

https://doi.org/10.1145/3372938.3372977

Botturi, L., & Stubbs, T. (Eds.). (2008). Handbook of Visual

Languages for Instructional Design: Theories and

Practices. IGI Global. https://doi.org/10.4018/978-1-

59904-729-4

Cabrera, O., Franch, X., & Marco, J. (2017). Ontology-based

context modeling in service-oriented computing: A

systematic mapping. Data & Knowledge Engineering,

110, 24–53. https://doi.org/10.1016/j.datak.2017.03.008

Cano-Benito, J., Cimmino, A., & García-Castro, R. (2021).

Toward the Ontological Modeling of Smart Contracts:

A Solidity Use Case. IEEE Access, 9, 140156–140172.

IEEE Access. https://doi.org/10.1109/ACCESS.

2021.3115577

Celik, D., & Magoulas, G. D. (2016). A Review, Timeline,

and Categorization of Learning Design Tools. In D. K.

W. Chiu, I. Marenzi, U. Nanni, M. Spaniol, & M.

Temperini (Eds.), Advances in Web-Based Learning –

ICWL 2016 (Vol. 10013, pp. 3–13). Springer

International Publishing. https://doi.org/10.1007/978-

3-319-47440-3_1

Chang, M., D’Aniello, G., Gaeta, M., Orciuoli, F.,

Sampson, D., & Simonelli, C. (2020). Building

Ontology-Driven Tutoring Models for Intelligent

Tutoring Systems Using Data Mining. IEEE Access, 8,

48151–48162. IEEE Access. https://doi.org/10.1109/

ACCESS.2020.2979281

Chatti, M. A., & Muslim, A. (2019). The PERLA

Framework: Blending Personalization and Learning

Analytics. The International Review of Research in

Open and Distributed Learning, 20(1).

https://doi.org/10.19173/irrodl.v20i1.3936

Chimalakonda, S., & Nori, K. V. (2020). An ontology based

modeling framework for design of educational

technologies. Smart Learning Environments, 7(1), 28.

https://doi.org/10.1186/s40561-020-00135-6

Demaidi, M. N., Gaber, M. M., & Filer, N. (2018).

OntoPeFeGe: Ontology-Based Personalized Feedback

Generator. IEEE Access, 6, 31644–31664. IEEE Access.

https://doi.org/10.1109/ACCESS.2018.2846398

Dresch, A., Lacerda, D. P., & Antunes Jr, J. A. V. (2015).

Design Science Research: A Method for Science and

Technology Advancement. Springer International

Publishing. https://doi.org/10.1007/978-3-319-07374-3

Eyal, L., & Gil, E. (2020). Design patterns for teaching in

academic settings in future learning spaces. British

Journal of Educational Technology, 51(4), 1061–1077.

https://doi.org/10.1111/bjet.12923

Ghazal, R., Malik, A. K., Qadeer, N., Raza, B., Shahid, A.

R., & Alquhayz, H. (2020). Intelligent Role-Based

Access Control Model and Framework Using Semantic

Business Roles in Multi-Domain Environments. IEEE

Access, 8, 12253–12267. IEEE Access.

https://doi.org/10.1109/ACCESS.2020.2965333

Gruber, T. R. (1993). A translation approach to portable

ontology specifications. Knowledge Acquisition, 5(2),

199–220. https://doi.org/10.1006/knac.1993.1008

Guarino, N., Oberle, D., & Staab, S. (2009). What Is an

Ontology? In S. Staab & R. Studer (Eds.), Handbook on

Ontologies (pp. 1–17). Springer Berlin Heidelberg.

https://doi.org/10.1007/978-3-540-92673-3_0

Hernández-de-Menéndez, M., Morales-Menendez, R.,

Escobar, C. A., & Ramírez Mendoza, R. A. (2022).

Learning analytics: State of the art. International Journal

on Interactive Design and Manufacturing (IJIDeM),

16(3), 1209–1230. https://doi.org/10.1007/s12008-022-

00930-0

Hyunsook, C., & Jeongmin, K. (2016). An Ontological

Approach for Semantic Modeling of Curriculum and

Syllabus in Higher Education. International Journal of

Information and Education Technology, 6(5), 365–369.

https://doi.org/10.7763/IJIET.2016.V6.715

Iatrellis, O., Kameas, A., & Fitsilis, P. (2019). EDUC8

ontology: Semantic modeling of multi-facet learning

pathways. Education and Information Technologies,

24(4), 2371–2390. https://doi.org/10.1007/s10639-019-

09877-4

Jang, S., & Woo, W. (2012). 5W1H: Unified User-Centric

Context. 2.

Katis, E., Kondylakis, H., Agathangelos, G., & Vassilakis,

K. (2018). Developing an Ontology for Curriculum and

Syllabus. In A. Gangemi, A. L. Gentile, A. G.

Nuzzolese, S. Rudolph, M. Maleshkova, H. Paulheim,

J. Z. Pan, & M. Alam (Eds.), The Semantic Web: ESWC

2018 Satellite Events (Vol. 11155, pp. 55–59). Springer

International Publishing. https://doi.org/10.1007/978-

3-319-98192-5_11

Koper, R., & Bennett, S. (2008). Learning Design:

Concepts. In H. H. Adelsberger, Kinshuk, J. M.

Pawlowski, & D. G. Sampson (Eds.), Handbook on

Information Technologies for Education and Training

(pp. 135–154). Springer Berlin Heidelberg.

https://doi.org/10.1007/978-3-540-74155-8_8

Lacasta, J., Lopez-Pellicer, F. J., Espejo-García, B.,

Nogueras-Iso, J., & Zarazaga-Soria, F. J. (2018).

Agricultural recommendation system for crop protection.

Computers and Electronics in Agriculture, 152, 82–89.

https://doi.org/10.1016/j.compag.2018.06.049

Lalingkar, A., Ramnathan, C., & Ramani, S. (2014).

Ontology-based smart learning environment for

teaching word problems in mathematics. Journal of

Computers in Education, 1(4), 313–334.

https://doi.org/10.1007/s40692-014-0020-z

LeClair, A., Jaskolka, J., MacCaull, W., & Khedri, R.

(2022). Architecture for ontology-supported multi-

context reasoning systems. Data & Knowledge

Engineering, 140, 102044. https://doi.org/10.1016/j.

datak.2022.102044

Lui, A. M., & Bonner, S. M. (2016). Preservice and

inservice teachers’ knowledge, beliefs, and

instructional planning in primary school mathematics.

Teaching and Teacher Education, 56, 1–13.

https://doi.org/10.1016/j.tate.2016.01.015

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

106

Mangaroska, K., & Giannakos, M. (2019). Learning

Analytics for Learning Design: A Systematic Literature

Review of Analytics-Driven Design to Enhance

Learning. IEEE Transactions on Learning

Technologies, 12(4), 516–534. https://doi.org/10.1109/

TLT.2018.2868673

Mizoguchi, R., Sinitsa, K., & Ikeda, M. (1996). Task

Ontology Design for Intelligent Educational/Training

Systems. Workshop on Architectures and Methods for

Designing Cost-Effective and Reusable.

Ouissem, B., Lamia, M., & Hafidi, M. (2021). A Proposed

Ontology-Based Generic Context Model for Ubiquitous

Learning. International Journal of Web-Based

Learning and Teaching Technologies, 16(3), 47–64.

https://doi.org/10.4018/IJWLTT.20210501.oa4

Ouyang, F., Wu, M., Zheng, L., Zhang, L., & Jiao, P. (2023).

Integration of artificial intelligence performance

prediction and learning analytics to improve student

learning in online engineering course. International

Journal of Educational Technology in Higher Education,

20(1), 4. https://doi.org/10.1186/s41239-022-00372-4

Peffers, K., Tuunanen, T., Rothenberger, M. A., &

Chatterjee, S. (2007). A Design Science Research

Methodology for Information Systems Research.

Journal of Management Information Systems, 24(3),

Article 3. https://doi.org/10.2753/MIS0742-1222240302

Pelap, G. F., Fitime, L. F., & Mbaiossoum, B. L. (2023).

Ontology-Based Modeling of the Learner in a Web

Educational System: Towards Learning Analytics and

Adaptive Learning. 8(7).

Perera, C., Zaslavsky, A., Christen, P., & Georgakopoulos,

D. (2014). Context Aware Computing for The Internet

of Things: A Survey. IEEE Communications Surveys &

Tutorials, 16(1), 414–454. https://doi.org/10.1109/

SURV.2013.042313.00197

Pozzi, F., Asensio-Perez, J. I., Ceregini, A., Dagnino, F. M.,

Dimitriadis, Y., & Earp, J. (2020). Supporting and

representing Learning Design with digital tools: In

between guidance and flexibility. Technology,

Pedagogy and Education, 29(1), 109–128.

https://doi.org/10.1080/1475939X.2020.1714708

Rahayu, N. W., Ferdiana, R., & Kusumawardani, S. S.

(2022). A systematic review of ontology use in E-

Learning recommender system. Computers and

Education: Artificial Intelligence, 3, 100047.

https://doi.org/10.1016/j.caeai.2022.100047

Ramaswami, G., Susnjak, T., Mathrani, A., & Umer, R.

(2023). Use of Predictive Analytics within Learning

Analytics Dashboards: A Review of Case Studies.

Technology, Knowledge and Learning, 28(3), 959–980.

https://doi.org/10.1007/s10758-022-09613-x

Ramesh, R., M., S., & Iyer, S. (2016). Integrating the

Learning Objectives and Syllabus into a Domain

Ontology for Data structures Course. Proceedings of

the 2016 ACM Conference on Innovation and

Technology in Computer Science Education, 266–271.

https://doi.org/10.1145/2899415.2899453

Reynolds, S., Pate, W. C., & Ochoa, O. (2023). An

Ontology and Management System for Learning

Outcomes and Student Mastery. 2023 IEEE Frontiers

in Education Conference (FIE), 1–5.

https://doi.org/10.1109/FIE58773.2023.10343171

Romero, C., & Ventura, S. (2020). Educational data mining

and learning analytics: An updated survey. WIREs Data

Mining and Knowledge Discovery, 10(3), e1355.

Romero, L., Link to external site, this link will open in a

new window, Saucedo, C., Ma, L. C., & Gutiérrez, M.

(2019). Supporting self-regulated learning and

personalization using ePortfolios: A semantic approach

based on learning paths. International Journal of

Educational Technology in Higher Education, 16(1),

1–16. https://doi.org/10.1186/s41239-019-0146-1

Rowley, J. (2007). The wisdom hierarchy: Representations

of the DIKW hierarchy. Journal of Information

Science, 33(2), 163–180.

Schmitz, M., Van Limbeek, E., Greller, W., Sloep, P., &

Drachsler, H. (2017). Opportunities and Challenges in

Using Learning Analytics in Learning Design. In É.

Lavoué, H. Drachsler, K. Verbert, J. Broisin, & M.

Pérez-Sanagustín (Eds.), Data Driven Approaches in

Digital Education (Vol. 10474, pp. 209–223). Springer

International Publishing. https://doi.org/10.1007/978-

3-319-66610-5_16

Susnjak, T., Ramaswami, G. S., & Mathrani, A. (2022).

Learning analytics dashboard: A tool for providing

actionable insights to learners. International Journal of

Educational Technology in Higher Education, 19(1),

12. https://doi.org/10.1186/s41239-021-00313-7

Tatto, M. T., Rodriguez, M. C., & Reckase, M. (2020).

Early career mathematics teachers: Concepts, methods,

and strategies for comparative international research.

Teaching and Teacher Education, 96, 103118.

https://doi.org/10.1016/j.tate.2020.103118

Verdú, E., Regueras, L. M., Gal, E., de Castro, J. P., Verdú,

M. J., & Kohen-Vacs, D. (2017). Integration of an

intelligent tutoring system in a course of computer

network design. Educational Technology Research and

Development, 65(3), 653–677. https://doi.org/10.1007/

s11423-016-9503-0

Vu, T. M. H., & Tchounikine, P. (2021). Supporting teacher

scripting with an ontological model of task-technique

content knowledge. Computers & Education, 163,

104098.

Vu, T.-M.-H., Le Dinh, T., Dam, N. A. K., & Pham-Nguyen,

C. (2023). Context-aware Knowledge-based Systems: A

Literature Review. https://hdl.handle.net/10125/103190

Wang, Y., & Wang, Y. (2021). A Survey of Ontologies and

Their Applications in e-Learning Environments.

Journal of Web Engineering, 20(6), 1675–1720.

Journal of Web Engineering. https://doi.org/10.13052/

jwe1540-9589.2061

Wei, N., & Shao, X. (2022). A Semantic Model of Internet

of Things for Intelligent Translation and Learning.

Mathematical Problems in Engineering,

2022.

https://doi.org/10.1155/2022/1651288

Wen, Y., Zhu, X., & Zhang, L. (2022). CQACD: A Concept

Question-Answering System for Intelligent Tutoring

Using a Domain Ontology With Rich Semantics. IEEE

Access, 10, 67247–67261. IEEE Access.

Support Learning Design and Analytics with EduP Knowledge Model

107