Aulakh Et Al - 2023 - E-Learning Enhancement Through Educational Data Mining With Covid-19 Outbreak

Since January 2020 Elsevier has created a COVID-19 resource centre with
free information in English and Mandarin on the novel coronavirus COVID-

19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company's public news and information website.
Elsevier hereby grants permission to make all its COVID-19-related

research that is available on the COVID-19 resource centre - including this
research content - immediately available in PubMed Central and other
publicly funded repositories, such as the WHO COVID database with rights
for unrestricted research re-use and analyses in any form or by any means
with acknowledgement of the original source. These permissions are
granted for free by Elsevier for as long as the COVID-19 resource centre
remains active.
International Journal of Educational Development 101 (2023) 102814
Contents lists available at ScienceDirect
International Journal of Educational Development

journal homepage: www.elsevier.com/locate/ijedudev
E-learning enhancement through educational data mining with Covid-19

outbreak period in backdrop: A review
Kudratdeep Aulakh *, Rajendra Kumar Roul, Manisha Kaushal
Thapar Institute of Engineering and Technology, Patiala, Punjab, India
A R T I C L E I N F O A B S T R A C T
Keywords: E-learning is fast becoming an integral part of the teaching- learning process, particularly after the outbreak of
Educational data mining Covid-19 pandemic. Educational institutions across the globe are striving to enhance their e-learning instruc
E-learning systems tional mechanism in accordance with the aspirations of present-day students who are widely using numerous
Educational data
technological tools — computers, tablets, mobiles, and Internet for educational purposes. In the wake of the
Student performance prediction
Covid-19 pandemic
evident incorporation of e-learning into the educational process, research related to the application of Educa
tional Data Mining (EDM) techniques for enhancing e-learning systems has gained significance in recent times.
The various data mining techniques applied by researchers to study hidden trends or patterns in educational data
can provide valuable insights for educational institutions in terms of making the learning process adaptive to
student needs. The insights can help the institutions achieve their ultimate goal of improving student academic
performance in technology-assisted learning systems of the modern world. This review paper aims to compre
hend EDM’s role in enhancing e-learning environments with reference to commonly-used techniques, along with
student performance prediction, the impact of Covid-19 pandemic on e-learning and priority e-learning focus
areas in the future.
Practitioner Notes • Educational institutions around the world have to recognize the
growing need to incorporate e-learning and other technology-
What is already known about this topic: assisted mechanisms for instructional purposes.
• Increased use of technology in education has necessitated academi
• Educational Data Mining techniques are applied by researchers to cians/administrators to restructure the teaching-learning process in
extract valuable information from educational data. tune with the changing times, with the help of new models proposed
• Application of EDM techniques can help educational institutions by EDM researchers.
improve e-learning/online learning systems. • New, advanced EDM techniques proposed by researchers, can help
• The Covid-19 pandemic has forced educational institutions to switch educational institutions make the learning process more adaptive to
to e-learning systems so that the learning process could continue the needs of the students and also support personalized learning.
without any interruption.
What this paper adds: 1. Introduction
• It provides an overview of selected past research papers related to
EDM applications to e-learning, including the backdrop of the Covid Educational Data Mining (EDM) is an evolving research domain that
pandemic period. underlies the concept of ‘Data Mining in Education’ (Grigorova et al.,
• It presents a review of past research pertaining to EDM use in e- 2017). It pivots around four key stakeholders — students, teachers,
learning from the perspective of student academic performance administrators, and researchers. EDM involves an extensive application
prediction. of Data Mining techniques to large volumes of data obtained from
• It highlights the growth of e-learning in recent years and the pro educational institutions (Shahiri et al., 2015). According to a definition
jections of increased significance of e-learning in the post-Covid era. given by The International EDM Society, and also by (Baker et al., 2010),
Implications for practice and/or policy: EDM is ‘an emerging discipline’ which is focused on ‘developing
* Corresponding author.
https://doi.org/10.1016/j.ijedudev.2023.102814
Received 19 November 2022; Received in revised form 20 March 2023; Accepted 22 April 2023
Available online 19 May 2023
0738-0593/© 2023 Elsevier Ltd. All rights reserved.
K. Aulakh et al. International Journal of Educational Development 101 (2023) 102814
methods for exploring the unique types of data that come from educa
tional settings,’ together with the use of the developed methods ‘to
better understand students and the settings which they learn in.’ The
EDM technique is aimed at the extraction of hidden but useful infor
mation from huge educational datasets pertaining to educational set
tings at all levels — from the school level to the university level
(Bhardwaj and Pal, 2012). EDM plays a very crucial role in ‘mining’ the
most precise information about the behavior of students as well as
gauging the efficacy of the learning process (Sana et al., 2019). The use
of EDM methodology reveals useful knowledge about the educational
settings, and facilitates the discovery of helpful trends and patterns from
large and complex educational datasets (Han et al., 2011). (Fig. 1).
The application of EDM by researchers to e-learning environments is
aimed at evaluating the teaching-learning process and suggesting im
provements to the process from time to time. The common EDM appli
Fig. 2. Growth of published papers on E-Learning system in educational
cation areas in e-learning include analyzing and visualizing educational institutions.
data, examining student behavior, predicting student performance,
learning-related suggestions for students, providing feedback to teach
a radical change in the educational sector across the globe.
ers, designing courseware, and scholastic planning, and laying out the
This review paper focuses on the need to improve the e-learning
study schedule (Romero and Ventura, 2010). Since the ultimate objec
method of education to align it appropriately with the ongoing era of
tive of educational institutions is to enhance existing education systems
rapid technological advancements, especially after the onslaught of the
through better decision-making capability of the administrators (Silva
pandemic which brought the teaching-learning process to an unprece
and Fonseca, 2017). The four prominent spheres to which EDM research
dented halt across the globe. Hence, the main objective of this review
efforts, from the e-learning perspective are presently being directed
paper is to understand how EDM can be applied to e-learning systems
include:
with focus on the ‘enhancement’ perspective, including the need for
improving the e-learning scenario against the backdrop of the pandemic
• Understanding the learning behaviors of students
(Azadi et al., 2021; Aqeel et al., 2022; Su et al., 2021; Abbas, 2020). The
• Evaluating or predicting student performance/grades
key research questions which comprised the backdrop of this objective
• Assessing student engagement and student satisfaction levels
were:
• Analyzing the reasons behind student drop-out rates.
• Can EDM application to e-learning improve the design of e-learning
Researchers have been showing an increasing interest in the e-
systems to better suit student needs?
learning scenario, which is evident from the growth of published papers
• How can student engagement and student performance be predicted
on e-learning, as shown in Fig. 2 which is based on the information
in e-learning systems?
drawn from ScienceDirect database. The reason why e-learning has been
• Has Covid-19 pandemic affected the approach of educational in
garnering the interest of EDM researchers is that the e-learning mech-
stitutions towards e-learning?
anism is being adopted by educational institutions over the past few
• What are the potential areas in e-learning on which educational in
years.
stitutions should focus in the post-Covid years?
The figure depicts the growth in published works in e-learning area
over the last two decades. The figure shows that published papers on e-
This review is significant because e-learning systems having gained
learning systems in educational institutions witnessed a more or less
prominence in recent times, particularly after the pandemic crisis.
continual growth from 1998 to 2015. Thereafter, from 2015 to 2018,
Hence, the most important differentiating aspect of this review, in
there was a decline in published works, seemingly because of a greater
comparison to other surveys, is the backdrop of the pandemic, owing to
focus of researchers on Artificial Intelligence and Blockchain. However,
which e-learning is projected to come to the forefront of education
after 2018, research related to e-learning systems in educational in
systems. The pandemic has forced educational institutions to scale up
stitutions started recording notable growth again. More explicitly, the
their online instructional mechanism so as to avoid interruptions to the
use of EDM techniques in e-learning systems has attracted renewed
learning process. With traditional classroom teaching having being
attention of researchers after Covid-19 pandemic, which brought about
Fig. 1. Overview of the abstract.
2
brought to an unexpected standstill by the pandemic, institutions are analysis of educational data can be beneficial for students, educators as
switching over to online teaching methods at an impressive pace, both in well as administrators because it can help them understand the learning
the developed and the developing countries. Hence, this review draws process and work together to improve it (Baker, 2015).
attention to the increased scope for application of EDM techniques to
improve e-learning systems for the future. 2.1.2. Sources and types of data used in EDM
The review paper is divided into different sections which focus on In e-learning systems, the educational data obtained by EDM re
EDM and e-learning, with the above-mentioned research questions in searchers for analysis is the data available from different sources
the backdrop. Section 2 of the paper gives an overview of EDM and e- (Romero et al., 2010), some of which have been shown in Fig. 4.
learning, and also highlights the diverse areas of application of EDM to In general, the diverse learning contexts which are generally
e-learning environments. Section 3 comprises the Literature Survey, considered important sources of data generation in e-learning are
with focus on the use of EDM approaches for student performance pre Computer-based education, Learning Management Systems, Intelligent
diction and the impact of the Covid-19 pandemic on e-learning envi Tutoring Systems, Interactive Education Settings, Adaptive Intelligent
ronment. Section 4 highlights the relevance of the review and its Hypermedia Systems, data-loaded learning activities like quiz, and
implications for e-learning systems. Section 5 comprises of discussion educational games. Data obtained from different learning contexts, as
section. Section 6 highlights the open issues in EDM. Section 7 comprises mentioned in Fig. 5, can be evaluated by EDM researchers to offer so
the conclusion of the paper and future research directions. lutions to questions pertaining to language, motivation, social discourse,
etc (Koedinger et al., 2015). This data can facilitate analysts, researchers
2. Educational data mining and e-learning: overview and academicians in improving e-learning environments so as to
enhance the teaching- learning process (Khanna et al., 2016). The
2.1. Educational data mining diverse aspects of e-learning process can contribute to the generation of
various types and categories of data.
2.1.1. The process of data mining in education The use of EDM techniques can help researchers in downsizing huge
EDM is a research area which has originated from Data Mining educational data into valuable information (Baritchi, 2004). By
(Anjewierden et al., 2007). It pivots around the educational setting, analyzing educational data, researchers can obtain helpful pattern
thereby marking a new application domain in Data Mining (DM) for s/trends from massive datasets (Bharati and Ramageri, 2010), discover
extracting knowledge from educational data, as depicted in Fig. 3. Be unique patterns which are formerly unknown (Qureshi et al., 2013), and
sides, knowledge extraction, EDM researchers are also concerned with spot unidentified patterns in student behavior (Janczewski and Colarik,
recommending improvements to existing educational systems (Via 2007). By using EDM techniques, researchers can offer potential solu
lardi-Sacín et al., 2009). EDM aims at augmenting student-learning tions to a number of education-related issues associated with prediction
models, which are typically indicative of the current knowledge of stu of student enrollment in a learning program, syllabus arrangement,
dents as well as their learning attitudes (Baker et al., 2009). forecasting student success with course material, identifying learning
EDM enables researchers to collect the data from diverse educational patterns, and understanding student behavior (Kumar and Chadha,
environments, analyze the data, understand the actions and behaviors 2011).
of students, and determine their academic success (Whitley, 2018).
Educational datasets are analyzed by EDM researchers with the help of 2.1.3. Educational data mining techniques
various techniques from numerous fields. In fact, EDM involves the EDM techniques are different methods which researchers use for
various disciplines of DM, including include Soft computing (Mitra and mining knowledge from educational datasets. These techniques often
Acharya, 2005), Machine learning (Witten et al., 2017), Statistics (Hill overlap with other areas like machine learning, neural networks, data
et al., 2006), Probability (Karegar et al., 2008), Natural language visualization, pattern recognition, and knowledge acquisition (Sumathi
(McCarthy and Boonthum-Denecke, 2011), and Artificial intelligence and Sivanandam, 2006). The application of different techniques to
(Bhattacharyya and Hazarika, 2006). Besides these disciplines, re educational data enables researchers to help institutions identify prior
searchers also use Database Systems and Recommender Systems ity learning requirements of various student groups, recognize at-risk
(Bousbia and Belamri, 2014). They apply various tools, techniques, al learners, analyze student and institutional performance, and opti
gorithms, and research designs to educational datasets and suggest mize learning resources (Algarni, 2016). The use of EDM techniques has
learning-related improvements to educational institutions. The main gained significance in recent years because the accountability of stu
orientation in EDM is towards the designing of methods, models, algo dents’ success largely lies with educational institutions (Campbell
rithms and tasks which can facilitate researchers in analyzing the data et al., 2007). The application of EDM techniques can play a vital role in
obtained from educational environments (Peña-Ayala, 2014). The transforming the current teaching and learning models (Brown et al.,
Fig. 3. ’Extraction of Knowledge’ component of EDM.
3
Fig. 4. Educational systems which contribute data for analysis.
Fig. 5. Types of data used for Educational Data Mining.
2015). The techniques facilitate the collection and analysis of educa teachers and administrators for the purpose of tracking students’ aca
tional data and assist the researchers in discovering previously undis demic progress and evaluating the effectiveness of different instruc
covered information, relationships and latest learning-related trends tional practices (Hamilton et al., 2009). EDM techniques can pave the
(ZoriÄ et al., 2020). By using EDM techniques, researchers can study the way for optimization of institutional proficiency, and enhance learning
learning process exhaustively and provide useful suggestions with re outcomes by offering a personalized, interactive and adaptive learning
gard to improvement of student performance, course selection by stu environment for students (Papamitsiou and Economides, 2014). EDM
dents, student retention, and regulation of the finances of an educational techniques can be a powerful method for assessing whether the learning
institution (Goyal and Vohra, 2012). material keeps the students interestingly engaged with the study course
There are a number of techniques which can be applied to educa (Cocea and Weibelzahl, 2009). The knowledge obtained by EDM re
tional data, depending on the pre-determined task of a particular data searchers via the use of various techniques can be beneficial for assessing
mining process. Commonly, there are four data mining tasks: Prediction, the academic progress of students, use of innovative teaching methods
to forecast student performance (Şen et al., 2012); Classification, to by teachers/instructors, and the education-centric decision-making
categorize data into fixed groups (Nisbet et al., 2009); Clustering, to process of academic planners (Jacob et al., 2015).
group together similar objects (Larose and Larose, 2014); and Associa
tion, to find interesting relationships amongst different data variables 2.2. E-learning
(Raval, 2012). In addition, the Regression technique is also widely used
by EDM researchers. Table 1 briefly summarizes the common EDM The term ‘e-learning,’ first used in the year 1998, refers to a
techniques and their application scenarios. technology-assisted learning system which is commonly known as Web-
EDM researchers can use one — or a combination — of techniques to based learning or Internet-based learning. It is a new-age learning
analyze educational datasets, depending on the research objective. The environment in which different types of learning materials and digital
choice of specific techniques is largely governed by the researchers’ technologies are used for providing a customized, learner-focused, and
standpoint as regards a particular educational aspect (Hanna, 2004). interactive educational environment to students (Rodrigues et al.,
The use of various techniques allows researchers to offer insights to 2019). The e-learning environment is, therefore, a combination of two
4
Table 1 groups (Sangrà et al., 2012):

EDM application scenerios and techniques used.
Educational Data Mining Application Scenario Technique Commonly • A learning system driven by technology
Used • A learning system involving communication and interactivity
Prediction of student performance/grades; Prediction • A learning system centered on delivery system
understanding student • A learning system oriented towards educational model.
behavior
• Decision Tree E-learning has, over the years, become a rapidly-progressing trend in
Algorithm
• Support Vector
education (Wang, 2003). It is one of the most important applications of
Machine digital technologies in learning environments (Rakic et al., 2020).
• Neural Network E-learning is the process of teaching and learning online via network
Categorization and profiling of students; finding out Classification technologies, marking an extremely powerful response to the
their learning
ever-increasing need for education (Zhang et al., 2004). E-learning
styles/preferences
• Decision trees systems offer a wide array of multimedia and interactive educational
• K-nearest neighbour resources which can contribute to the designing of suitable educational
• Statistical analysis environments for students in various situations (Haughey et al., 2008;
• Bayesian classification Bates, 2008; Cleveland-Innes and Garrison, 2010).The e-learning mode
Rule-based methods
of education delivery can yield almost the same learning outcomes —
•
• Neural Networks
• Memory-based generally assessed in grades or marks – for students as the traditional
reasoning classroom system (Russell, 2001). Overall, e-learning can make the
• Support vector educational processes quicker and more effective than traditional
machines
classroom-teaching methods.
Grouping of students according to knowledge and Clustering
content-
interaction patterns; grouping of similar course 2.2.1. Role of teachers in e-learning environments
materials The digital era has brought along rapid progress in the use of infor
• K-means mation and communications technologies, due to which there has been a
• K-medoids
momentous transformation in education systems around the globe
• Probabilistic
clustering (Arinto, 2016). The emergence of the e-learning concept has brought
• Relocation algorithm about a notable change in the role of teachers as compared to the
• Agglomerative traditional classroom-learning systems. The responsibility of teachers
algorithm
has evolved from that of being merely a dispenser of knowledge to that
• Divisive algorithm
• Density function of being the curator of innovative spaces in which students can explore
Clustering knowledge (Siemens, 2008). In contrast to the previous educational
• Density-based method of teacher-centric knowledge transmission, the new education
Connectivity systems focus on the student-centric knowledge generation method
Study of relationships in student behavior patterns; Relationship Mining
(Armellini and Jones, 2008; Swan, 2020). This is so because e-learning
identifying problems in teaching • Casual Data Mining
• Association rule environments involve widespread availability of massive amounts of
mining information which underline a vast range of student-teacher interactions
• Sequential Pattern that are continually generated in an educational setting (Abd Elaal,
Mining
2013). As such, there is greater emphasis on the role of teachers in terms
• Correlation Mining
Prediction of numeric values (continuous values) in a Regression
of the design of learning environments conducive to the needs of stu
given dataset, for analysis of dependent-variable and • Linear Regression dents (Anderson, 2004; Bennett et al., 2009).
independent-variable relationship • Logistic Regression The present-generation students are able to actively benefit from e-
• Polynomial learning environments because of their capability to interact with
Regression
different forms of technology for diverse purposes (Vai and Sosulski,
2011; Ko and Rossen, 2017; Sahoo et al., 2020). Therefore, owing to the
key concepts (i) learning; and (ii) technology. E-learning is one of expansive use of e-learning systems of late, teachers have to undertake a
several technology-assisted learning systems — besides virtual learning, big responsibility in study-course development which needs to be
m-learning etc. — in which the utilization of computerized methods is reformulated in accordance with use of technology in education
conceptualized for the objective of facilitating the teaching-learning (Abrioux, 2001). Teachers have to adapt their teaching methods to
process (Aparicio et al., 2016). The e-learning methodology facilitates modern-day e-learning environments in which the learning courses are
numerous types of interactions and conversations which can enable quite prominently based on resource (Naidu, 2007) or online dialogue
close association between teachers and students (Calvert, 2005). (Jara and Mohamad, 2007). Teachers are expected to play a more pro
E-learning is aimed at offering a configurable educational network in active role in the e-learning environment, so that digital-education
which learning materials, learning instruments and instructional ser services can be improved and the reach of these services can be
vices are amalgamated into one distinct solution for creating and expanded to all students in the future (Falakmasir and Habibi, 2010).
delivering educational or training content in a fast, effective and The engagement of students with e-learning materials offered by
economical manner (Ong et al., 2004). In the e-learning system of ed educational institutions has become a crucial fac- tor for improving the
ucation, knowledge is available to all learners without any incongruity chances of students’ academic success. The advancements in technology
linked to time restrictions or geographical location (Sun et al., 2008). In have enabled teachers to manage online courses for learners by using
fact, the biggest benefit of e-learning is that the technology-assisted online platforms such as Google Classroom, Udemy, etc. With digital
learning models ensure that the constraints of time and space do not channels making it possible for teachers to reach study materials to their
impact the student-to-student interactions and student-to-teacher students, the use of Learning Management Systems has become quite
communication (Katz, 2002). Based on its different definitions, the common. The use of online platforms and other technologies by teachers
concept of e-learning can broadly be categorized into the following have given students the advantage of accessing the learning materials
from any place, on any Internet-connected device, such as personal
5
computer, laptop, smartphone, etc. This online access to study materials EDM methodology to e-learning includes some specific tasks, such as
online has emerged as a real savior for education systems during the evaluation of students’ learning performance,
Covid-19 pandemic. personalizing student learning via course adjustments and learning
suggestions, assessing online-learning content, and developing suitable
2.2.2. Recent trends in E-learning models to understand learning behavior of students in e-learning courses
The emergence of diverse e-learning scenarios and the rapid (Castro et al., 2007).
expansion of e-learning content have marked the advent of new From the functional perspective, some distinct areas of application of
technology-based e-learning trends, as illustrated in Fig. 6. Some of EDM to e-learning environments include identification of student com
these trends, such as mobile learning, are already being incorporated in plaints, analysis of student drop-out rates, and comprehensive assess
the e-learning process. However, a few new trends, like the use of ment of student and teacher performance (Ali, 2013). Meanwhile, with
Augmented and Virtual Reality, are somewhat futuristic trends which reference to higher education, the main application areas of EDM are:
are expected to gain ground in the coming years. In addition, new student performance forecasting, evaluation of student learning, and
technologies like the Internet of Things (IoT) also mark a significant facilitating teaching and research work (Ray and Saeed, 2018). Broadly
technological advancement in the field of e-learning (Soni, 2019). There speaking, there are four key EDM application areas in e-learning (Prabha
is great focus on how IoT can be applied to e-learning (Priatna et al., and Shanavas, 2014), as illustrated in Fig. 7.
2020), and also on leveraging IoT for bringing about an improvement in On the whole, there are wide-ranging applications of EDM in e-
learning outcomes for students (Kassab et al., 2020). Besides IoT, learning for improving the teaching-learning process in tune with the
another technology set for wide application in e-learning is cloud needs of the students.
computing which will apparently enhance the efficiency of e-learning
process due to the cloud-storage advantage (Malhi et al., 2020). Artifi 3. Literature review
cial Intelligence is another modern-day technology which can contribute
immensely to the transition to e-learning (El Gourari et al., 2020) and to To compile the literature review of EDM application to e-learning,
a more personalized learning system for students, especially in the wake some relevant past research papers published between 2015 and 2020
of the Covid pandemic. were selected. However, a few papers from previous years and early
The latest trends in e-learning are largely aimed at identification of 2021 have also been included because of their significance to the topic
the learning-content space, development of new e-learning content and and its implications for e-learning. The words included in search query
personalization of learning content for new-generation learners. were: ‘Application of educational data mining to e-learning’; ‘Educa
tional data mining techniques and e-learning’; ‘Educational data mining
2.3. Application of educational data mining to e-learning and student performance prediction in e-learning;’ and ‘online learning
during Covid-19 pandemic.’.
Researchers can apply various EDM techniques to e-learning sce The literature survey has been divided into three sections, based on
narios to analyze educational datasets from the perspective of enhancing the grouping of relevant published papers. The first section includes
the educational process in the digital age. From a practical view, re papers related to use of EDM in e-learning in general; the second section
searchers can use valuable information from student-learning data for includes papers related to EDM use for student performance prediction
assessing an educational setting, recommending qualitative improve in e-learning systems; and the third section includes papers related to
ments to the instructional systems, and laying the foundation for an study of impact of Covid-19 pandemic on e-learning environment. The
enhanced learning process (Romero et al., 2004). The application of criterion for selection and grouping of the papers for first two sections
Fig. 6. Recent E-learning trends.
6
Fig. 7. Key areas of Educational Data Mining application in E-learning.
was that the papers either involved the use of different EDM techniques learning platform were studied by (Zhang and Qin, 2018), who elabo
to e- learning or presented some specific approaches to improve e- rated how raw educational data can be transformed into knowledge with
learning systems. The papers reviewed in the third. the help of EDM techniques.
section mainly pertain to an analysis of e-learning environment The integration of mobile learning methodology into e-learning
during Covid-19 pandemic. system was studied by (El-Sofany and El-Haggar, 2020) to focus on the
effectiveness of mobile learning methods in e-learning environments
for improving the learning outcomes of students.
3.1. Use of educational data mining methodology in e-learning Technology-assisted self-directed learning was discussed by (Alti
environment mari et al., 2012), with focus on learning objects and integration of
repeat-use content units into personalized learning paths of students.
The scope of resolution of diverse education-centric problems E-learning and student engagement levels in different higher-
through the use of EDM to educational data was surveyed by (Ganesh education online learning platforms were studied by (Grubišić et al.,
and Christy, 2015). 2020) to track student learning activities and student knowledge.
EDM techniques were applied by (Gushchina and Ochepovsky, 2020) In brief, the research papers reviewed in this section highlight the
to an e-learning system with the intention of understanding the behavior fact that although EDM is a fairly new research domain, the efforts of
of students and their interest level with educational content. researchers to analyze educational data for improving e-learning sys
An e-learning system to overcome the challenges related to e- tems have been quite significant. The popularity of e-learning systems
learning platforms has been proposed by (Appalla et al., 2017), based on has increased manifold in a short time span because of the ‘adaptively’
the common challenges hampering the effectiveness of e-learning sys advantage which has transformed the learning process for students.
tems and delivering personalized educational materials to students. There has also been a vast expansion of student- information databases
An assessment of student engagement in a technology-mediated because of a remarkable growth in the availability of online education
education system was attempted by (Henrie et al., 2015) with the help resources. Researchers can mine valuable knowledge from educational
of different student-engagement evaluation measures, like surveys and datasets and facilitate the decision-making process of educational
other quantitative self-report methods, as well as observational institutions.
measures.
A case study for enhancing e-learning systems was presented by
(Jovanovic et al., 2012), who used data mining on educational datasets 3.2. Application of educational data mining techniques to predict student
pertaining to student behavior and cognitive styles. performance
The effectiveness of web-based courses and e-learning systems for
improving overall student-learning experiences was studied by (Ashraf In e-learning related research, an attempt was made by (Daghestani
and Khan, 2017) to highlight that e-learning systems can enhance the et al., 2020) to gauge the impact of learning materials with student
self-learning ability of students. engagement, by drawing attention to the concept of Artificial Intelli
An understanding of e-learning from the standpoint of time conti gence for providing new prospects to make e-learning more effective.
nuity as a distinct characteristic was attempted by (Wan and Niu, 2019), A new approach for the application of EDM to understand student
who spotlighted the acute scarcity of information faced by e-learning performance in e-learning environments has been presented by (Injadat
systems. et al., 2020), who studied the use of multiple techniques for forecasting
The application of EDM for evaluating the behavior of students using student performance to support their academic progress.
online learning platforms has been discussed by (Jie et al., 2017), who EDM has been described by (Rawat and Malhan, 2019) as a practice
used EDM techniques and algorithms to analyze the study behavior of in which useful information is automatically obtained and seg- mented
students based on data related to students’ landing behavior and from the educational data sources, and algorithms are used for
resources-explored behavior. student-performance prediction.
The core concepts of EDM, Big Data in Education, and online A study of use of supervised data mining techniques to understand
7
students’ learning environment interactions and their performance in Fast-paced developments in the EDM and Learning Analytics have
exams was conducted by (Tomasevic et al., 2020). been highlighted by (Kausar et al., 2018) who focused on integration of
Different approaches and applications of EDM were studied by (Kaur EDM techniques in a personalized e-learning system, and underlined the
and Bathla, 2018) from the context of prediction of student performance need to obtain valuable information in an adaptive manner from
after analyzing hidden trends and patterns in voluminous educational educational data.
datasets. The application of Recommender systems in e-learning was
Table 2
Summary of EDM techniques/approaches presented.
Name of the paper Reference No. Purpose Technique/Approach Outcome
Recommender system for predicting (Thai-Nghe et al., To predict the performance of Application of Recommender Better student- performance prediction
student performance 2010) students in e-learning scenario systems in e-learning outcomes as compared to conventional
regression techniques
Authoring tutored, adaptive e- courses (Altimari et al., 2012) To study how Learning Objects Proposed an adaptive, learner- Technology-assisted learning can be
in a personal learning environment: a can be changed into content centric Personal Learning enhanced by dynamically integrating
dynamic syllabus and dynamic units for repeated usage for Environment (PerLE) reusable content units into students’
assembly approach students personalized learning paths
Using data mining on student behavior (Jovanovic et al., To analyze student behavior Clustering, Classification Focus on preprocessing of data,
and cognitive style data for improving 2012) and cognitive styles selection of attributes, and optimization
e- learning systems: a case study of parameters can improve student
performance
An analysis of the determinants of (Castillo-Merino and To study the key factors Using data from an Motivation is the most significant
students’ performance in e-learning Serradell-López, affecting student performance experimental system variable affecting student performance
2014) in e-learning systems established as part of eLene-EE and learning efficiency in e-learning set-
project up
Applications of educational data (Ganesh and Christy, To identify algorithms with Evaluation of clustering and Suggested potential directions for EDM
mining: a survey 2015) high reliability for future EDM classification algorithms in future
research
An efficient educational data mining (Appalla et al., 2017) To understand reasons for in- Proposed e-learning systems Proposed e-learning systems overcame
approach to support e-learning effectiveness of e-learning using five knowledge-input challenges pertaining to e-learning
systems steps platforms
Effectiveness of Data Mining (Ashraf and Khan, To study how data mining Intelligent Adaptive system Improvement in real-life e-learning
Approaches to E- Learning System: A 2017) approaches that contribute to environment
Survey enhancement of e-learning
environ-ment
Application of educational data mining (Jie et al., 2017) To evaluate behavior of Study of log data generated in Teaching content should be orga- nized
on analysis of students’ online students using online learning specifically designed according to students’ study behavior
learning behavior platforms Autonomous Learning Platform for op- timizing teaching strategies
Student performance prediction using (Kaur and Bathla, To implement EDM techniques Naive Bayes and Support Extensive use of DM algorithms is
educational data mining techniques 2018) for predicting student Vector Machine techniques crucial for the purpose of decision-
performance and learning making and student anal-ysis from
behaviors diverse aspects
Integration of data mining clustering ap- (Kausar et al., 2018) To highlight fast- paced Approach for integrating Identification of the best educational
proach in the personalized E-learning developments in EDM and Clustering technique in settings for students to enhance their
system Learning Analytics fields personalized e-learning system learning capabilities
A brief analysis of the key technologies (Zhang and Qin, To introduce core concepts of Classification of important data Laid out directions for future research
and applications of educational data 2018) EDM, Big Data in Education, mining technologies as Big Data can have a substantial
mining on online learning platform and online learning platform impact on online education
An evaluation of student performance at (Rakic et al., 2019) To explore the assessment of K-means Clustering, Linear Students’ success is hugely impacted by
e-learning platform student performance at e- Regression, and Social Network digital learning content at e-learning
learning platforms Analysis platforms
A hybrid e-learning recommendation (Wan and Niu, 2019) To focus on time continuity and Proposed Hybrid Filtering Personalized and diversified recom-
approach based on learners’ influence acute scarcity of information recommendation approach mendations can make e-learning
propagation faced by e-learning systems systems more efficient and adaptable to
student needs
Adapting gamified learning systems (Daghestani et al., To gauge the impact of learning Proposed an Adaptive Gamified Artificial Intelligence (AI) can make e-
using educational data mining 2020) materials with student Learning System (AGLS) learning more effective by increasing
techniques engagement student engagement and learning
performance
The effectiveness of using mobile (El-Sofany and To gauge the effectiveness of Integration of mobile-learning Mobile-learning methods offer greater
learning techniques to improve El-Haggar, 2020) use of mobile learning methods methodology into e-learning flexibility to students for accessing
learning outcomes in higher edu- in e-learning environments system educational content
cation
Data mining of students’ behavior in E- (Gushchina and To understand students’ Cluster analysis method, V-fold Gauged student interest levels, evalu-
learning system Ochepovsky, 2020) learning behavior cross-checking, data analysis ated quality of learning materials
and visualization
A common model for tracking student (Grubišić et al., 2020) To track student learning Proposed a model aimed at Evaluated student engagement levels
learning and knowledge acquisition in activities and student Tracking Student Learning and with e-learning courses used for higher
different e-Learning platforms knowledge Knowledge (TSLAK) education
Systematic ensemble model selection (Injadat et al., 2020) To understand student KNN, Random Forest, SVM, More accurate prediction of student
approach for educational data mining performance in e-learning Naïve Bayes, LR and Multi- performance for supporting their
environments Layer Perception academic progress
An overview and comparison of (Tomasevic et al., To study students’ learning Model-based approach, Probabilistic approach Predicting
supervised data mining techniques for 2020) environment interactions and Similarity- based approach students’ exam performance and
student exam performance prediction their performance in exams forecasting dropout risk and future
achievements of students
8
considered by (Thai-Nghe et al., 2010) with focus on recommendation of underlined the crucial role played by e-learning infrastructure for
books, papers, and other educational materials to students in an making distance learning more effective during the pandemic. In addi
e-learning environment. tion, (Wang et al., 2021) highlighted sustainable development goals in
The need for assessment of student performance in e-learning plat education due to the pandemic impact; (Zarei and Mohammadi, 2021)
forms has been highlighted by (Rakic et al., 2019) who listed the benefits studied e-learning infrastructure from the perspective of developed and
of e-learning platforms in terms of flexibility in teaching as well as wide developing countries; (Qiao et al., 2021) analyzed how online learning
availability of learning content. programs are being developed/adopted by institutions as a result of
The key factors that affect student performance in e-learning systems Covid; and (Rajab et al., 2020) stressed on the need to make e-learning
have been underlined by (Castillo-Merino and Serradell-López, 2014) systems more effective by extending the reach of technology to students.
who said that student performance depends on several variables, A tabular summary of the review of published papers related to study
including the abilities of students, teaching methods and technologies of e-learning systems during the Covid pandemic is presented in Table 3.
used by educators, and technology adoption level of institutions. With the onset of the COVID-19 pandemic, healthcare systems
To summarize, these two sections of literature survey show that re around the world have been under immense strain. This pandemic has
searchers have focused on student-performance prediction by ascer had an impact on all aspects of life.
taining interconnected features that affect performance. The survey Some other papers included in the study investigates how contin
underscores the need for defining, evaluating and comparing student gencies disturb existing theoretical models and their consequences for
academic performance in e-learning systems and conventional online purchases in the post-COVID-19 age. During the COVID-19 crisis,
classroom-teaching methods. The review also shows that widespread use customers prefer to use applications on websites for search and purchase
of e-learning systems has necessitated the application of more accurate (Al Halbusi et al., 2022; Moradi et al., 2021).
mechanisms for dynamically predicting student performance. The paper rebuilds and constructs the revival, experts, professionals,
On the basis of the literature survey of published papers included in and academia must evaluate changes in the fundamental principles,
Sections 3.1 and 3.2, a summary of EDM methods used, or approaches primary assumptions, and business organisations’ situations related to
presented, by researchers is given in Table 2. the service industry and practise framework (Fu et al., 2021).
The table shows that the academic performance of students is usually To accomplish sustainable education, social media plays an essential
predicted by researchers with the help of EDM techniques like Classi part in the learning behaviour of university students. The impact of
fication (Naive Bayes, K-Nearest Neighbor, Random Forest); Clustering social media on long-term schooling is becoming an important and
(K-means); Prediction (Support Vector Machine, Decision tree); motivating component (Abbas et al., 2019).
Regression and Relationship Mining. These techniques are widely The usage of the internet and social media technology has arisen as
applied by researchers to analyze e-learning environments (Silva et al., an essential instrument of human civilization, and the evolution of
2021); understand learning behavior (Aguinis et al., 2021), academic technological integration, cyberspace, and web-technology has become
performance (Wiliam and Thompson, 2008) and learning styles of standard practise in educational institutions. Online use among students
students (Dwivedi et al., 2019) and provide feedback to the e-learning has become crucial in learning behavior; yet, excessive internet and
stakeholders (Shvets et al., 2020). social media use leads to internet addiction (Khazaie et al., 2021).
The first step in understanding how the COVID-19 pandemic has
3.3. | Review of e-learning environment during Covid-19 pandemic used the funding resources already available and what kind of spending
would be required to successfully prepare for the next pandemic is to
A global perception of e-learning scenario during lockdowns due to identify how much has been spent in the health sector as a result of the
Covid-19 pandemic has been presented by (Radha et al., 2020) with pandemic. In order to comprehend the financial background of the
emphasis on improved self-learning skills of students especially those pandemic and the relative scale of shift in expenditure that would be
familiar with web technology. required to effectively avert the next global pandemic, it is important to
Online education has been described by (Dhawan, 2020) as a look at global health spending at the beginning of the pandemic (Micah
panacea for learning during Covid-19 pandemic period because the et al., 2023).
pandemic has resulted in a scenerio in which providing online education The emergence of the COVID-19 epidemic has sparked a global crisis
to students is no longer an option for educational institutions; it has that has forced the closure of educational institutions all over the planet.
become a necessity. This study intends to investigate the factors that influence students’
An affirmation that Internet played a key role to support the intentions to use e-library services during the tough COVID-19 pandemic
teaching-learning processes during the pandemic crisis has made by (Rahmat et al., 2022).
(Favale et al., 2020) who stressed the availability of Internet connec On account of the review, it is evident that instructional dimensions
tivity was crucial for supporting the teaching- learning processes during are set for a notable change in future, after the pandemic. On one hand,
the pandemic. it would be crucial for educational institutions to adopt new technolo
The level of student satisfaction in e-learning systems during the gies and provide an excellent digital framework for education while, on
pandemic has been studied by (Saxena et al., 2021) who exam- ined the the other hand, students will have to gain proficiency in the use of
impact of the crisis on the quality of e-learning processes based on fac learning-technology tools — like computers, mobile devices, Internet
tors like Internet content, responsiveness, assurance, and reliability. etc. — and also learn to efficiently interact with these tools. However, it
The need to develop effective e-learning systems against the back is important that universal and affordable access to the Internet should
drop of Covid-19 pandemic has been underlined by (Ananga, 2020) who be available to students so that they can benefit from technology-
noted that the rapidly increasing use of Internet for education world assisted education environments.
wide has put the spotlight on demand and supply of learning materials.
In some of the other papers pertaining to e-learning review during 4. Relevance of the review and implications for e-learning
Covid-19 pandemic, (Yildiz et al., 2020) highlighted that the transition
to online and distance education has, by and large, had a positive impact 4.1. Relevance of the review
on students. Meanwhile,
Nikou and Maslov (2021) focused on perceived utility and conve This review paper is relevant in present times because it focuses on
nience of e-learning systems; (Tawafak et al., 2021) emphasized on the need to improve the e-learning method of education to align it
improved behavior intention of students for using e-learning technolo appropriately with the ongoing era of fast-paced technological de
gies because of the pandemic impact; and (Garad et al., 2021) velopments. The relevance of the review can more specifically be
9
Table 3 Table 3 (continued )

Analysis of e-learning environment during Covid-19 pandemic. Name of Paper Reference No. Observations Views on e-
Name of Paper Reference No. Observations Views on e- learning
learning enhancement
enhancement
cognitive improving grants from
E-Learning during (Radha et al., Pandemic made e- E-learning can competence in distance- learning government,
lock-down of 2020) learning an in- improve self- distance learning effectiveness teachers’
Covid-19 tegral part of learning skills of effectiveness during pandemic trainings and
pandemic: A school/college/ students and their during the covid- workshops etc.
global perspec- university academic 19 pandemic
tive education performance Impact of COVID-19 (Wang et al., Pandemic forced E-learning
Online learning: A (Dhawan, Educational Changed on achieving the 2021) educational systems can
panacea in the 2020) institutions have educational goal of sus- institutions world- improve with
time of COVID-19 shifted focus from environment tainable wide to set up e- enhanced focus
crisis traditional brings new development: E- learning systems on students,
classroom teaching opportunities and learning and on war footing teachers, study
methods to challenges for educational plan, technology,
technology- improved e- productivity and envi-
assisted education learning systems ronment
Campus traffic and e- (Favale et al., Internet played a Internet Challenges of higher (Zarei and E-learning marked Developed
Learning during 2020) vital role in robustness is education related Mohammadi, a comprehensive countries have
COVID-19 supporting important for to e-learning in 2021) response of the more enhanced e-
pandemic educational successfully developing education system learning
process during management of e- countries during to the pandemic infrastructure
pandemic learning process COVID-19 spread: than developing
Pedagogical (Ananga, Demand and Institutions are a review of the countries
Consider- ations of 2020) supply of widely exploring perspectives of
E-Learning in educational and adopting students,
Education for content has improved e- instructors, policy
Development in changed due to learning meth- makers, and ICT
the Face of COVID- increased internet ods, especially in experts
19 use for learning higher education The Development (Qiao et al., Institutions can Evolution and
Current Trends in (Yildiz et al., Educational Enhanced e- and Adoption of 2021) transition from in- adoption of
Education 2020) institutions rapidly learning systems Online Learning in class to online technology can
Technologies shifted to e- can positively Pre-and Post- education through foster e-learning
Research learning mode due impact student COVID-19: Combi- widespread use of environments for
Worldwide: Meta- to pandemic learning and nation of e-learning tools continuing
Analysis of Studies student Technological and technologies education during
between 2015 and satisfaction System Evolution and after
2020 Theory and pandemic
Challenges to online (Rajab et al., E-learningsystems Technology- Unified Theory of
medicaleducation 2020) should be made a related Acceptance and
during the COVID- more effective challenges and Use of Technology
19 pandemic mechanism for obstacles should
education be overcome to
increase explained as follows:
effectiveness of e-
learning systems
• The paper acknowledges the noteworthy rise in the adoption of e-
Examining the (Saxena et al., Pandemic affected E-learning quality
moderating effect 2021) student- can be improved learning systems by educational institutions worldwide in recent
of perceived satisfaction levels by focusing on years. The amalgamation of technology and educational content has
benefits of in e-learning Internet content, paved the way for institutions to switch over to e-learning and other
maintaining social systems receptiveness, technology-assisted mechanisms for instructional purposes.
distance on e- dependability,
learning quality and assurance
• Increased use of technology in 21st century has made it pertinent for
during COVID-19 educational systems to restructure their institutional mechanisms in
pandemic tune with the changing times. For this, educational institutions
An analysis of stu- (Nikou and Pandemic Efforts are depend greatly on researchers who can put forth new technology-
dents’ perspectives Maslov, 2021) awareness and required to make
assisted education models.
on e-learning challenges have e- learning
participation–the directly affected systems more • EDM researchers in the past have proposed enhancements to e-
case of COVID-19 student intention useful and more learning systems. They have been using various techniques for
pandemic in e-learning adaptive to mining valuable information from educational datasets, in order to
systems student needs understand and evaluate student behavior and student performance
Impact of (Tawafak Extensive use of Enhanced e-
Technologies et al., 2021) technology-aided learning
in e-learning settings.
During COVID-19 e- learning systems environments • The paper points to the fact that the pandemic has dramatically
Pandemic for during pandemic should enable changed the educational environments all over the world. It has
Improving Be- students to disrupted the normal course of instructional mechanisms in educa
havior Intention to develop
tional institutions.
Use E-learning technology-
related skills for • The prolonged impact of the pandemic would prompt EDM re
better academic searchers to study the use of new, advanced techniques that can
performance help educational institutions enhance their e-learning system and
The role of e- (Garad et al., E-learning Improved e- make the learning process more adaptive to student needs.
learning in- 2021) infrastructure learning systems
frastructure and played vital role in require financial
The relevance of the review can also be highlighted by the fact that
constructive dialogue, talks and discussions related to e-learning
10
systems in developed countries gained ground in 2020 and 2021, as is Table 5

evident from Table 4 and Table 5. The tables, based on information E-learning and training conferences scheduled in 2021.
obtained from LearnDash and Class Central sources, show the number of Name of the Conference Start date End Date Location/
e-learning related conferences that were scheduled in US and Europe Virtual
during these two years following the Covid pandemic. Future of Education Technology January 26 January 29 Virtual
Conference Virtual
Learning Leaders Online Forum February 24 February 26 Virtual
4.2. | Implications for e-learning systems Open Education Week 2021 March 1 March 5 Virtual
World Congress Virtual Educa March 3 March 5 Virtual
Learnit Live: The Future of March 8 March 12 Virtual
EDM is one of the prominently emerging research fields of data Learning is Now
mining, which has a proven role in education. Effective usage of online SXSW EDU March 9 March 12 Virtual
learning can improve the quality of educational resources while also Rutgers Online Learning March 15 March 16 Virtual
allowing educators to plan lessons and students to widen their horizons. Conference
OLC Innovate 2021 Virtual March 15 March 19 Virtual
It enables educators to examine a student’s particular needs and adjust Conference
classes to meet them. One significant advantage of using the internet for SXSW Conference March 16 March 20 Virtual
education is that it makes classes available even in remote settlements Empower Webinar Week April 6 April 8 Virtual
and rural locations where girls frequently experience educational dis 11th International Learning April 12 April 16 Virtual
Analytics and Knowledge
parities. Furthermore, the usage of the internet by students allowed
Conference
them to continue their educational endeavors during the Coronavirus EduTECH EUROPE 2021 April 28 April 29 Virtual
outbreak, which brought the world to a halt. Learning Solution Conference May 4 May 6 Orlando, FL
This review paper points to some of the major EDM application areas and Expo
in e-learning over the past few years. It underlines the application of Educase Learning Initiative May 18 May 20 Virtual
Annual Meeting
EDM techniques in the e-learning systems to achieve two broad objec
Qatar Digital Learning Summit June 2 June 3 Doha+Virtual
tives of e-learning: REMOTE: The Connected June 9 June 11 Virtual
(i) improving the quality of the teaching-learning process and (ii) Faculty Summit by ASU
ensuring that the learning process meets the needs of students. The London edtech week June 14 June 21 London
The Learning Ideas Conference June 16 June 18 Virtual
Covid-19 has unsettled the global education industry, which has been hit
Learning @ Scale 2021/ June 22 June 25 Virtual
really hard because of lockdowns. As a result of the pandemic, there has EMOOCs 2021
been a complete transformation of the education sector to an online ICDE 2021 June 22 June 23 Rome
mechanism globally. The uncertainties in education, attributable to the ISTE Digital Leadership Summit June 27 June 30 San Antonio
pandemic, have highlighted the significance and trustworthiness of e- Distance Teaching and Learning July 30 August 5 Virtual
Conference
learning methods. These methods are projected to expand further in the
ASU+GSV Summit: The Dawn of August 9 August 11 San Deigo
post-Covid years, owing to the projected development of new means of the Age of Digital Learning
using digital technology for its effective integration into student ALT Annual Conference 2021 September September Virtual
education. 7 9
16th European Conference on September September Bozen-Bolzano
Hence, the main implication of this review for EDM research and e-
Technology Enhanced 20 24
learning is that it identifies some potential priority areas in e-learning, as Learning
illustrated in Table 6. Learning with MOOCs VIII September October 1 Antigua
29 Guatemala
Open Education Global 2021 October 5 October 7 Nantes, France
Table 4
EADTU I-HE2021 November 3 November 5 Bari
E-Learning and training Conferences scheduled in 2020. Innovate Learning 2021 Summit November 9 November Virtual
Name of the Conference Start date End Date Location 11
WEEF/GEDC 2021 November November Madrid
In the USA 15 18
Curation and Learning Lab February 4 February 6 Saratoga Springs, OEB Berlin December 1 December 3 Berlin
NY 2021 International Conference December December Singapore
TechKnowledge Conference February 5 February 7 San Jose, CA on Technologies in STEM 14 15
ITC 2020 Annual eLearning February 9 February 12 Charleston, SC ’LIVE’
Conference
Training 2020 Conference February 24 February 26 Lake Buena Vista,
and Expo FL
The above-mentioned priority areas should be explored by EDM re
Learning Solutions and March 31 April 2 Orlando, FL
Realities 360 searchers to help educational institutions make their e-learning pro
OLC Innovate March 31 April 3 Chicago IL grams more effective in future.
ADT2020 May 17 May 20 Denver, CO
Learning DevChamp 2020 June 2 June 5 Salt Lake City, UT
Training Industry Conference June 16 June 18 Raleigh, NC 4.3. Broad application areas of EDM
and Expo
WPCampus July 15 July 17 New Orleans, LA There are numerous applications of EDM in e-learning for improving
TechLearn Conference 2020 October 13 October 15 Austin, TX
DevLearn October 21 October 23 Las Vegas, NV
the teaching-learning process to create a better environment for meeting
WorldCamp US October 27 October 29 St. Louis, MO the students’ requirements. Some of the well-recognized application
Slate Conference November November Naperville, IL areas of EDM are:
11 13
OLC Accelerate November November Lake Buena Vista,
1. Data analysis and visualisation: Data analytics is the act of
17 20 FL
In Europe evaluating data sets to make decisions about the information they
Learning Technologies February 12 February 12 London, UK have, whereas data visualisation is the pictorial or graphical
World of Learning Summit May 15 May 15 London, UK depiction of information and data.
WorldCamp Europe June 4 June 6 Porto, Portugal 2. Providing feedback for supporting instructors: Feedback is an
World of Learning Summit October 13 October 14 Birmingham, UK
essential component of the continuous learning and assessment
11
Table 6 from learner data, which is typically obtained by an intelligent

Potential areas for EDM application to E-learning in future. tutoring system as a result of the learner’s engagement with the
Name of Paper Author Suggestions for improving e- tutoring system.
learning environment 6. Supporting the creation of student groups: Grouping students
Exploring factors that (King and Focus on infrastructure of refers to the employment of several methods to organise pupils in
influence adoption of e- Boyatt, 2015) educational institutions, the classroom. They are usually organised to work in different
learning within higher skills/attitudes of staff- ways during each lesson. The teacher frequently chooses group
education members, and other factors ings based on the activity type, the activity’s goal, and the pupils.
which can influence e-learning
adoption
7. Adapting learning contents, sequences, and interfaces:
An exploratory case study of (Wolfinger, Re-determining the role of Adaptive learning is a teaching and learning system that aims to
middle school student 2016) teachers in facilitating fully tailor classes, readings, practice exercises, and assessments to
academic achievement in a online virtual learning for individual students based on their existing abilities and
fully online virtual school school students to improve
performances.
student performance
A recommender model in e- (Bourkoukou Devising new learning models 8. Social network analysis: Social network analysis is the practice
learning environment et al., 2017) to support personalized of examining social systems using networks and graph theory. It
learning experiences for defines networked systems in terms of nodes and the ties, edges,
improving student or links that connect them. Social network analysis techniques
performance and
achievements
are used to examine patterns of relationships among people in
Educational Data Mining: A (Rodrigues Need to develop teaching- groups. They are useful for evaluating persons or organisations’
review of evaluation process et al., 2018) learning scenarios and other social structure and interdependencies (or work patterns).
in the e-learning learning activities by focusing 9. Studying relationships between teachers and learners: A
on behavioral research,
positive relationship between the teacher and the student in the
interaction, collaboration, and
performance classroom is defined as an endeavour to develop trust and respect
Time really matters: (Tang et al., Important to shift to a new for each other. This relationship could include getting to know
Understanding the temporal 2019) viewpoint on online teaching- the students better, giving them options, and motivating them to
dimension of online learning process, by analyzing become better daily learners.
learning using educational inter-related learning events
10. Supporting teachers in concept map creation: Concept maps
data mining
Online IS education for the (He et al., 2014) Need to study the training of are especially beneficial for visually learning students. Concept
21st century teachers in designing, maps are an effective study approach because they help one to
developing, teaching, and grasp the big picture: by beginning with higher-level concepts,
assessing online courses
they allow one to chunk material based on significant
Analysis of educational data (Ahuja et al., Understanding the challenges
mining 2019) (like accuracy, functionality, relationships.
and overheads, etc.) linked to 11. Assisting construction and reuse of learning content: The
extremely large educational knowledge structure is fostered by the learner’s self- regulation
datasets and self-awareness. As a result, one must assist learners in
Evaluation of Blackboard (Alturise, 2020) Focus on potential
acquiring important skills and information and provide tools and
Learning Management improvements in the
System for Full Online application of e-learning circumstances to enhance their ability to control their learning.
Courses in Western Branch systems to meet the changing 12. Enhancing educational planning and scheduling: Educational
Colleges of Qassim needs of students planning is a process in which each student engages in self-
University
assessment, investigating and integrating academic and profes
Exploring challenges of online (Pandit and Need to create effective e-
education in COVID times Agrawal, 2021) learning ecosystem by
sional options, and making personal decisions for the present and
focusing on collaborative future, whereas educational scheduling organizes students ac
student learning, skill in using tivities throughout the day, week, term or year.
digital technologies, 13. Learning Management System: A learning management system
knowledge of online-
is a software application that is used to manage educational
education tools, academic
guidance etc. courses, training programmes, materials, or learning and devel
Making the Transition to (Stoetzel and Need to re-think and re-design opment activities. The concept of a learning management system
Virtual Methods in the Shedrow, 2021) lessons and assignments to evolved straight from e-learning.
Literacy Classroom: integrate technology into
14. Management of dropouts and retention of students: Dropout
Reframing Teacher learning materials to cope
Education Practices with unprecedented or
is defined as the termination of a programme, regardless of the
unforeseen situations reason, whereas retention refers to the student’s continued
participation in the programme after the original completion
date.
process. Providing frequent and continual feedback to the in
structors is an important way to improve learning achievement. Diverse EDM application areas can allow administrators and acade
3. Providing recommendations to students: Recommender sys micians to use data mining tools and participate in activities that can
tems are used to produce personalised curricula, as well as to make data collection and processing more accessible to all EDM users.
suggest individualised content on e-learning platforms, and it can Examples include statistical and visualization tools that analyze social
also be used to forecast student achievement before they enrol for networks and their influence on learning outcomes and productivity.
any higher educations.
4. Student performance prediction: Predicting student perfor 5. Discussion
mance is essential, which aids in course selection and establishing
appropriate future study programmes. This review of EDM application to e-learning coincides with the
5. Learner modelling: A learner model, also known as a student widespread adoption of e-learning systems over the past few years, more
model, is an organised depiction of the knowledge, mis specifically after the outbreak of the Covid-19 pandemic (Yu et al., 2022;
conceptions, and challenges of a learner. Learner models are built Su et al., 2022; Shoib et al., 2022; Zhou et al., 2022).There is no denying
12
that e-learning has garnered unprecedented attention after the EDM use in e-learning, specifically in terms of using some other database
pandemic, which had an unsettling impact on almost all industries or complex search techniques involving the application of algorithms.
worldwide, with the education industry having been hit really hard. The Moreover, this review has also left the option open for similar reviews
pandemic led to a complete transformation of the education sector to an involving applying different techniques/concepts by researchers to
online mechanism globally. It has prompted academic planners to suggest enhancements to e-learning for future-generation students,
restructure the traditional classroom-learning systems and prepare for a particularly with extensive use of technology and the pandemic impact
much-required shift to smart education systems that involve increased in retrospection.
use of technology tools and the Internet.
Against that backdrop, this review paper attempts to provide a 6. Open issues in educational data mining
general overview of how researchers have applied EDM techniques to
propose improvements to e-learning systems in the past few years. This The main objective of EDM research is to understand the factors
paper also highlights the need to upscale research efforts in this domain which affect student learning in the modern-day online learning or e-
to meet the needs of the present-generation and future-generation stu learning systems. EDM researchers seek to understand student behaviors
dents. The research papers selected for the review were divided into the and conceptualize the new-age learning environments that can endorse
following three groups to carry out a meaningful literature survey the reinforcement of the education process outside the traditional
related to the use of EDM for improving e-learning systems — papers classrooms. With researchers chiefly aimed at potential improvements in
related to the application of EDM to e-learning in general, papers asso teaching-learning methods, there are several open issues in EDM. Some
ciated with EDM use for student performance prediction in e-learning; of the most prominent issues are:
and papers related to the study of the impact of Covid-19 pandemic on
the e-learning environment. The criteria for selecting articles for the first • EDM is an iterative procedure of extracting useful information from
two groups were that the articles either involved using different EDM large educational datasets — it involves the use of repetitive pro
techniques in the context of e-learning or presented some specific ap cesses of hypothesis formulation, testing of hypothesis, and fine-
proaches to improve e-learning systems. The research papers selected tuning of results. Hence, for greater credibility of research in EDM,
for the third group mainly focused on analyzing the e-learning envi the iterative process of data mining should be illustrated by re
ronment during the Covid-19 pandemic. They included some pertinent searchers in detail.
suggestions for improving e-learning systems in the future so that the • It is difficult for EDM researchers to identify the extent to which the
teaching-learning process should not be hampered due to unforeseen results of their studies pertaining to a specific learning scenario can
situations. be effectively transferred to other educational settings. Therefore,
This review paper focused on answering the research questions in the the contexts of research linked to data mining in education necessi
Introduction section. The research questions pertain to four important tate a comprehensive explanation by researchers.
spheres concerning EDM use in e-learning — enhancement of e-learning • The application of conventional data mining algorithms to educa
systems, student performance/engagement prediction, the impact of the tional attributes is not always possible because EDM researchers
pandemic on e-learning, and potential e-learning focus areas post- have to distill huge amounts of data which is a result of rapid data
pandemic. Based on the outcome of the literature survey, it can be growth in educational environments. Consequently, EDM research
stated that, in answer to the first research question, researchers have requires the use of more complex sets of algorithms rather than
applied EDM techniques to suggest improvements to the design of e- conventional algorithms.
learning systems and to suit student needs better. Moreover, concerning • In the case of educational findings, there is no mechanism for
the second research question, it is evident that student engagement and gauging the quality of insights provided by EDM researchers because
student performance in e-learning systems can be predicted by re they specifically try to discover the underlying models in educational
searchers using various techniques, such as classification, clustering, data. As such, it is important that EDM researchers should clearly
regression, and relationship mining, among several others. Concerning delineate an educational-data model as a representation of actual
the third research question, it can be said that the Covid-19 pandemic educational settings.
has profoundly impacted the educational institutions’ approach toward • There is an apparent perception that the anonymity of students can
e-learning (Iorember et al., 2022; Yao et al., 2022).Finally, in answer to be compromised when cumulative summation of student data within
the fourth research question, the review enlists some potential areas in an algorithm is allowed by educational institutions. It is, therefore,
e-learning on which educational institutions should direct their atten essential to change the perceptions linked to the understanding of
tion in the post-Covid years. student-identifying attributes associated with EDM research.
However, despite the review paper’s focus on EDM application to e-
learning from the perspective of e-learning improvement, including past Overall, it is important to point out that the open issues in EDM can
efforts and future potential, the review has certain limitations. The study be addressed in the future with the help of cross-disciplinary ap
presents only a general idea of EDM use in e-learning through a survey of proaches, instead of focusing on one particular specialized approach.
some selected past research papers; it is not an all-inclusive effort. The The use of different approaches can encompass various areas of
main thought that served as the basis for paper selection was that the proficiency and enhance the data-analysis process for EDM researchers.
selected papers essentially involved the use of EDM techniques in e- The open issues in EDM need to be addressed in a far-sighted manner, so
learning, as anticipated from their title, abstract, keywords, or intro that EDM research can help educational systems successfully overcome
duction. No specific search algorithms or techniques were used for paper the challenges posed by fast-paced technological advancements in the
selection; only certain search queries related to EDM for e-learning were 21st century.
used. For the selection of papers, the Google Scholar database was used.
In addition, the main aim was to keep the study specific from the 7. Conclusion and future research directions
standpoint of a selection of papers pertaining only to the application of
EDM techniques for enhancing e-learning. It underlined a conscious 7.1. Conclusion
effort to keep other techniques and concepts — like Deep Learning,
Machine Learning, Artificial Intelligence, Big Data, Learning Analytics, The review of EDM application to e-learning systems in this paper
etc. — out of the scope of the review. This conscious decision behind the has highlighted that EDM is a prominently emerging research field that
review has made it relevant for the present and future. The review can pivots around the use of diverse techniques to analyze educational
serve as a basis for more systematized and detailed surveys related to datasets and develop new educational models. The introduction of
13
interactive e-learning methods has unfolded numerous opportunities for

EDM to understand and evaluate the technology-assisted learning pro
cess in diverse contexts. The research-based understanding can subse
quently enable academic planners to develop ‘smart’ and efficient
educational systems which can support students as well as teachers.
This study has presented growth of published papers on e-learning
over the last two decades, reviewed the process of data mining in ed
ucation and also provided insights into the role of teachers in e-learning
environment. In the present work, a comprehensive review of various
EDM techniques based on the methodologies such as support vector
machines, Naïve Bayes, K-means clustering, linear regression, cluster
analysis, random forest, and multi-layer perceptron has been presented.
According to the methodology-based taxonomy, the review presented
the purpose, outcome, and technique of each problem and, along with
them, presented a review and analysis of EDM. This systematic analysis
has helped in understanding different methodologies for EDM so that
the most suitable one can be selected.
This review coincides with the outbreak of Covid-19 pandemic, Fig. 8. Global E-learning market size.
which has impacted the education sector like never before. The crisis is
said to have emerged as a ‘constructive disruptor’ for the educational • Analyzing student engagement and student satisfaction with e-
domain because it has augmented the role of technology in learning learning programs
while disrupting classroom teaching. The pandemic has provided a • Predicting student performance in fully online virtual learning
unique opportunity for EDM researchers to focus on adapting the environments
existing education system to futuristic learning. Owing to the pandemic, • Suggesting approaches that can make e-learning programs more
academic planners already have efforts underway to restructure the value-oriented
traditional classroom- learning structure and make a gradual shift to e- • Evaluating the possible increase in first-time students accessing on
learning systems. The present study has also explored EDM applications line programs in the post-Covid period.
in the current era. Furthermore, this research has also presented impli
cations for e-learning systems and reviewed recent trends in e-learning To summarize, future research would aim at the use of EDM in e-
based on previous research results. learning to identify students with access to online learning after the
Lastly the articles summarize and discusses the future research di Covid pandemic, propose models for more flexible study programs,
rections and challenges based on the previous research. Overall the recognize the need for student support, and explore the prospects of the
proposed research is expected to help researchers understand the limi realistic combination of classroom education and e-learning
tations and advantages of various EDM techniques and can help in environments.
contributing towards future development of methodologies related to e-
learning system. Ethical approval
Based on the review, it is projected that research pertaining to EDM
application to e-learning is bound to increase in the coming years. The This article does not contain any studies with human participants or
global e-learning market is poised for steady growth in the post-Covid animals performed by any of the authors.
period, as indicated in Fig. 7. The figure, based on the information
drawn from E-learning industry statistics and Research And Markets. Funding details
com, shows that the size of the global e-learning market has increased
tremendously since 2010 — from $32.1 billion in 2010, $107 billion in There is no funding for this research work.
2015, and approximately $250 billion in 2020. It is further set for an
exponential increase over the next 4–5 years and is expected to reach Conflict of interest
upto $325 billion in 2025.
Finally, this review also underlines the anticipation that developers The authors declare that they have no conflict of interest.
of online education platforms are likely to venture further into the
‘Smart Education’ arena after the pandemic. They are expected to Informed consent
launch improved e-learning features to enhance their platforms in line
with the students’ learning needs. (Fig. 8). Informed consent was obtained from all individual participants
included in the study.
7.2. Future research directions Author statement
The future direction for research in the field of EDM application to e- All persons who meet authorship criteria are listed as authors, and all
learning would be to focus on the emerging aspects and factors that may authors certify that they have participated sufficiently in the work to
impact the dynamic e-learning environment. With technology set to play take public responsibility for the content, including participation in the
a big role in education, particularly after Covid, it would be appropriate concept, design, analysis, writing, or revision of the manuscript.
to undertake research efforts to make the learning process more ‘need
aligned’ for students and help them achieve greater academic success. References
Some of the e-learning aspects which would necessitate the attention of
EDM researchers in the future will be: Abbas, J., 2020. The impact of coronavirus (SARS-CoV2) epidemic on individuals mental
health: the protective measures of Pakistan in managing and sustaining transmissible
disease. Psychiatr. Danub. 32 (3–4), 472–477.
• Proposing re-design of lessons and assignments to integrate tech
nology into the learning process
14
Abbas, J., Aman, J., Nurunnabi, M., Bano, S., 2019. The impact of social media on Cocea, M., Weibelzahl, S., 2009. Log file analysis for disengagement detection in e-
learning behavior for sustainable education: Evidence of students from selected Learning environments. Use Model. Use -Adapt. Interact. 19 (4), 341–385.
universities in Pakistan. Sustainability 11 (6), 1683. Daghestani, L.F., Ibrahim, L.F., Al-Towirgi, R.S., Salman, H.A., 2020. Adapting gamified
Abd Elaal, S.A.E., 2013. E-learning using data mining. 21th Int. Conf. Chin. Res. J. learning systems using educational data mining techniques. Comput. Appl. Eng.
Abrioux, D., 2001. Converting to online course and program delivery: Global Educ. 28 (3), 568–589.
perspectives. Int. Rev. Res. Open Distrib. Learn. 1 (2). Dhawan, S., 2020. Online learning: a panacea in the time of COVID-19 crisis. J. Educ.
Aguinis, H., Yu, L., Tosun, C., 2021. How to enhance scholarly impact: recommendations Technol. Syst. 49 (1), 5–22.
for university administrators, researchers and educators. Int. J. Contemp. Hosp. Dwivedi A., Dwivedi P., Bobek S., Zabukovšek S.S. Factors affecting students’
Manag. engagement with online content in blended. learning. Kybernetes 2019;.
Ahuja, R., Jha, A., Maurya, R., Srivastava, R., 2019. Analysis of educational data mining. El Gourari A., Skouri M., Raoufi M., Ouatik F. The Future of the Transition to E-learning
Harmony Search and Nature Inspired Optimization Algorithms. Springer,, and Distance Learning Using Artificial Intelligence. In: 2020 Sixth International
pp. 897–907. Conference on e-Learning (econf) IEEE; 2020. p. 279–284.
Al Halbusi, H., Al-Sulaiti, K., Abbas, J., Al-Sulaiti, I., 2022. Assessing factors influencing El-Sofany H., El-Haggar N. The effectiveness of using mobile learning techniques to
technology adoption for online purchasing amid COVID-19 in Qatar: moderating improve learning outcomes in higher education 2020;.
role of word of mouth. Front. Environ. Sci. 1039. Falakmasir M.H., Habibi J. Using educational data mining methods to study the impact
Algarni, A., 2016. Data mining in education. Int. J. Adv. Comput. Sci. Appl. 7 (6), of virtual classroom in e-learning. In: Educational Data Mining 2010 ERIC; 2010.
456–461. Favale, T., Soro, F., Trevisan, M., Drago, I., Mellia, M., 2020. Campus traffic and e-
Ali, M.M., 2013. Role of data mining in education sector. Int. J. Comput. Sci. Mob. Learning during COVID-19 pandemic. Comput. Netw. 176, 107290.
Comput. 2 (4), 374–383. Fu, Q., Abbas, J., Sultan, S., 2021. Reset the industry redux through corporate social
Altimari F., Plastina A.F., Cronin M.D., Servidio R., Caria M., Pedrazzoli A. Authoring responsibility: the COVID-19 tourism impact on hospitality firms through business
tutored, adaptive e-courses in a personal learning environment: a dynamic syllabus model innovation. Front. Psychol. 6686.
and dynamic assembly approach. In: Proceedings of the World Congress on Ganesh S.H., Christy A.J. Applications of educational data mining: a survey. In: 2015
Engineering and Computer Science, vol. 1; 2012. p. 236–242. International Conference on Innovations in Information, Embedded and
Alturise, F., 2020. Evaluation of Blackboard Learning Management System for Full Communication Systems (ICIIECS) IEEE; 2015. p. 1–6.
Online Courses in Western Branch Colleges of Qassim University. Int. J. Emerg. Garad, A., Al-Ansi, A.M., Qamari, I.N., 2021. The role of e-learning infrastructure and
Technol. Learn. (iJET) 15 (15), 33–51. cognitive competence in distance learning effectiveness during the covid-19
Ananga, P., 2020. Pedagogical Considerations of E-Learning in Education for pandemic. J. Cakrawala Pendidik. 40 (1), 81–91.
Development in the Face of COVID-19. Interna- tional J. Technol. Educ. Sci. 4 (4), Goyal, M., Vohra, R., 2012. Applications of data mining in higher education. Int. J.
310–321. Comput. Sci. Issues (IJCSI) 9 (2), 113.
Anderson, T., 2004. Teaching in an online learning context. Theory Pract. Online Learn. Grigorova K., Malysheva E., Bobrovskiy S. Application of data mining and process mining
273. approaches for improving e- learning processes. In: 3rd International Conference on
Anjewierden A., Kolloffel B., Hulshof C. Towards educational data mining: Using data Information Technology and Nanotechnology; 2017. p. 25–27.
mining methods for automated chat analysis to understand and support inquiry Grubišić, A., Žitko, B., Stankov, S., Šarić-Grgić, I., Gašpar, A., Tomaš, S., et al., 2020.
learning processes. In: International Workshop on Applying Data Mining in e- A common model for tracking student learning and knowledge acquisition in
Learning (ADML 2007) Citeseer; 2007. p. 27–36. different e-Learning platforms. J. e-Learn. Knowl. Soc. 16 (3), 10–23.
Aparicio M., Bacao F., Oliveira T. An e-learning theoretical framework. An e-learning Gushchina, O., Ochepovsky, A., 2020. Data Mining Of Students’ Behavior in E-Learning
theoretical framework 2016;(1):292–307. System. In: Journal of Physics: Conference Series, vol. 1553. IOP Publishing,,
Appalla, P., Kuthadi, V.M., Marwala, T., 2017. An efficient educational data mining 012027.
approach to support e-learning. Wirel. Netw. 23 (4), 1011–1024. Hamilton L., Halverson R., Jackson S.S., Mandinach E., Supovitz J.A., Wayman J.C.
Aqeel, M., Rehna, T., Shuja, K.H., Abbas, J., 2022. Comparison of students’ mental Using Student Achievement Data to Support Instructional Decision Making. IES
wellbeing, anxiety, depression, and quality of life during COVID-19’s full and partial Practice Guide. NCEE 2009–4067. National Center for Education Evaluation and
(smart) lockdowns: a follow-up study at a 5-month interval. Front. Psychiatry 13. Regional Assistance 2009;.
Arinto, P.B., 2016. Issues and challenges in open and distance e-learning: Perspectives Han, J., Pei, J., Kamber, M., 2011. Data mining: concepts and techniques. Elsevier,.
from the Philippines. Int. Rev. Res. Open Distrib. Learn. 17 (2), 162–180. Hanna, M., 2004. Data mining in the e-learning domain. Campus-wide Inf. Syst.
Armellini A., Jones S. Carpe Diem: seizing each day to foster change in e-learning design. Haughey, M., Evans, T., Murphy, D., 2008. Introduction: from correspondence to virtual
2008;. learning environments. Int. Handb. Distance Educ. 1–24.
Ashraf, A., Khan, M.G., 2017. Effectiveness of data mining approaches to E-learning He, W., Xu, G., Kruck, S., Online, I.S., 2014. education for the 21st century. J. Inf. Syst.
system: A survey. NFC IEFR J. Eng. Sci. Res. 4. Educ. 25 (2), 101–106.
Azadi, N.A., Ziapour, A., Lebni, J.Y., Irandoost, S.F., Abbas, J., Chaboksavar, F., 2021. Henrie, C.R., Halverson, L.R., Graham, C.R., 2015. Measuring student engagement in
The effect of education based on health belief model on promoting preventive technology-mediated learning: a review. Comput. Educ. 90, 36–53.
behaviors of hypertensive disease in staff of the Iran University of Medical Sciences. Hill, T., Lewicki, P., Lewicki, P., 2006. Statistics: methods and applications: a
Arch. Public Health 79 (1), 69. comprehensive reference for science, industry, and data mining. StatSoft, Inc.
Baker, R., et al., 2010. Data mining for education. Int. Encycl. Educ. 7 (3), 112–118. Injadat, M., Moubayed, A., Nassif, A.B., Shami, A., 2020. Systematic ensemble model
Baker, R., 2015. Big data and education. New York: Teachers College. Columbia selection approach for educational data mining. Knowl. -Based Syst. 200, 105992.
University,. Iorember, P.T., Iormom, B., Jato, T.P., Abbas, J., 2022. Understanding the bearable link
Baker, R.S., Yacef, K., et al., 2009. The state of educational data mining in 2009: A review between ecology and health outcomes: the criticality of human capital development
and future visions. J. Educ. - tional data Min. 1 (1), 3–17. and energy use. Heliyon, e12611.
Baritchi A. Data mining and knowledge discovery. In: Business Intelligence in the Digital Jacob J., Jha K., Kotak P., Puthran S. Educational data mining techniques and their
Economy: Opportunities, Limitations and Risks IGI Global; 2004.p. 35–47. applications. In: 2015 International Conference on Green Computing and Internet of
Bates, T., 2008. Transforming distance education through new technologies. Int. Handb. Things (ICGCIoT) IEEE; 2015. p. 1344–1348.
Distance Educ. 217–236. Janczewski L., Colarik A. Cyber warfare and cyber terrorism. IGI Global; 2007.
Bennett, S., Agostinho, S., Lockyer, L., Harper, B., 2009. Researching Learning Design in Jara M., Mohamad F. Pedagogical templates for e-learning. WLE Centre, Institute of
Open, Distance, and flExible Learning: In- Vestigating Approaches to Supporting Education, University of London; 2007.
Design Processes and Practices. Taylor & Francis,. Jie W., Hai-yan L., Biao C., Yuan Z. Application of educational data mining on analysis of
Bharati M., Ramageri M. Data mining techniques and applications 2010;. students’ online learning behavior. In: 2017 2nd International Conference on Image,
Bhardwaj B.K., Pal S. Data Mining: A prediction for performance improvement using Vision and Computing (ICIVC) IEEE; 2017. p. 1011–1015.
classification. arXiv preprint arXiv:12013418 2012;. Jovanovic, M., Vukicevic, M., Milovanovic, M., Minovic, M., 2012. Using data mining on
Bhattacharyya, D.K., Hazarika, S.M., 2006. Networks, Data Mining, and Artificial student behavior and cognitive style data for improving e-learning systems: a case
Intelligence: Trends and Future Directions. Narosa Pub House. study. Int. J. Comput. Intell. Syst. 5 (3), 597–610.
Bourkoukou, O., El Bachari, E., El, Adnani, M., 2017. A recommender model in e- Karegar M., Isazadeh A., Fartash F., Saderi T., Navin A.H. Data-mining by probability-
learning environment. Arab. J. Sci. Eng. 42 (2), 607–617. based patterns. In: ITI 2008–30th International Conference on Information
Bousbia, N., Belamri, I., 2014. Which Contribution Does EDM Provide to Computer- Technology Interfaces IEEE; 2008. p. 353–360.
Based Learning Environments? Educational Data Mining. Springer, pp. 3–28. Kassab, M., DeFranco, J., Laplante, P., 2020. A systematic literature review on Internet of
Brown, S., White, S., Power, N., 2015. Tracking undergraduate student achievement in a things in education: Benefits and challenges. J. Comput. Assist. Learn. 36 (2),
first-year physiology course using a cluster analysis approach. Adv. Physiol. Educ. 39 115–127.
(4), 278–282. Katz, Y.J., 2002. Attitudes affecting college students’ preferences for distance learning.
Calvert, J., 2005. Distance education at the crossroads. Distance Educ. 26 (2), 227–238. J. Comput. Assist. Learn. 18 (1), 2–9.
Campbell, J.P., DeBlois, P.B., Oblinger, D.G., 2007. Academic analytics: a new tool for a Kaur, H., Bathla, E.G., 2018. Student performance prediction using educational data
new era. EDUCAUSE Rev. 42 (4), 40. mining techniques. Int. J. Future Revolut. Comput. Sci. Commun. Eng. 4 (12), 93–97.
Castillo-Merino, D., Serradell-López, E., 2014. An analysis of the determinants of Kausar, S., Huahu, X., Hussain, I., Wenhao, Z., Zahid, M., 2018. Integration of data
students’ performance in e-learning. Com. - Puters Hum. Behav. 30, 476–484. mining clustering approach in the personalized E-learning system. IEEE Access 6,
Castro, F., Vellido, A., Nebot, A., Mugica, F., 2007. Applying data mining techniques to e- 72724–72734.
learning problems. Evolution of Teaching and Learning Paradigms in Intelligent Khanna L., Singh S.N., Alam M. Educational data mining and its role in determining
Environment. Springer,, pp. 183–221. factors affecting students academic performance: A systematic review. In: 2016 1st
Cleveland-Innes, M.F., Garrison, D.R., 2010. An Introduction to Distance Education: India international conference on information processing (IICIP) IEEE; 2016. p. 1–7.
Understanding Teaching and Learning in a New Era. Routledge.
15
Khazaie, H., Lebni, J.Y., Abbas, J., Mahaki, B., Chaboksavar, F., Kianipour, N., et al., Romero, C., Ventura, S., De, Bra, P., 2004. Knowledge discovery with genetic
2021. Internet addiction status and related factors among medical students: a cross- programming for providing feedback to courseware authors. Use Model. Use -Adapt.
sectional study in Western Iran. Int. Q. Community Health Educ. p. Interact. 14 (5), 425–464.
0272684X211025438. Romero C., Ventura S., Pechenizkiy M., Baker R.S. Handbook of educational data mining.
King, E., Boyatt, R., 2015. Exploring factors that influence adoption of e-learning within CRC press; 2010.
higher education. Br. J. Educ. Technol. 46 (6), 1272–1280. Russell, T.L., 2001. The no significant difference phenomenon: a comparative research
Ko, S., Rossen, S., 2017. Teaching Online: A Practical Guide. Routledge. annotated bibliography on technology for distance education. AL: Int. Distance Educ.
Koedinger K.R., D’Mello S., McLaughlin E.A., Pardos Z.A., Rose C.P. Data mining and Certif. Cent.
education. Wiley Interdisciplinary. Reviews: Cognitive Science 2015;6(4):333–353. Sahoo S., et al. E-readiness and perception of student teachers’ towards online learning in
Kumar, V., Chadha, A., 2011. An empirical study of the applications of data mining the midst of COVID-19 pandemic. Available at SSRN 3666914 2020;.
techniques in higher education. Int. J. Adv. Comput. Sci. Appl. 2 (3). Sana B., Siddiqui I.F., Arain Q.A. Analyzing students’ academic performance through
Larose, D.T., Larose, C.D., 2014. Discovering Knowledge in Data: an Introduction to Data educational data mining 2019.
Mining, vol. 4. John Wiley & Sons,. Sangrà, A., Vlachopoulos, D., Cabrera, N., 2012. Building an inclusive definition of e-
Malhi, M.S., Iqbal, U., Nabi, M.M., Malhi, M.A.I., 2020. E-learning based on cloud learning: An approach to the conceptual framework. Int. Rev. Res. Open Distrib.
computing for educational institution: Security issues and solutions. Int. J. Electron. Learn. 13 (2), 145–159.
Inf. Eng. 12 (4), 162–169. Saxena, C., Baber, H., Kumar, P., 2021. Examining the moderating effect of perceived
McCarthy P., Boonthum-Denecke C. Applied natural language: identification, benefits of maintaining social distance on e-learning quality during COVID-19
investigation and resolution. Information Science Reference, Pennsylvania 2011;. pandemic. J. Educ. Technol. Syst. 49 (4), 532–554.
Micah, A.E., Bhangdia, K., Cogswell, I.E., Lasher, D., Lidral-Porter, B., Maddison, E.R., Şen, B., Uçar, E., Delen, D., 2012. Predicting and analyzing secondary education
et al., 2023. Global investments in pandemic preparedness and COVID-19: placement-test scores: A data mining approach. Expert Syst. Appl. 39 (10),
development assistance and domestic spending on health between 1990 and 2026. 9468–9476.
Lancet Glob. Health 11 (3), e385–e413. Shahiri, A.M., Husain, W., et al., 2015. A review on predicting student’s performance
Mitra, S., Acharya, T., 2005. Data Mining: Multimedia, Soft Computing, and using data mining techniques. Procedia Comput. Sci. 72, 414–422.
Bioinformatics. John Wiley & Sons,. Shoib, S., Buitrago, J.G., Shuja, K., Aqeel, M., de Filippis, R., Abbas, J., et al., 2022.
Moradi, F., Ziapour, A., Abbas, J., Najafi, S., Rezaeian, S., Faraji, O., et al., 2021. Suicidal behavior sociocultural factors in developing countries during COVID-19.
Comparing the associated factors on lifestyle between type 2 diabetic patients and L’encephale 48 (1), 78–82.
healthy people: a case-control study. Int. Q. Community Health Educ. p. Shvets O., Murtazin K., Piho G. Providing Feedback for Students in E-Learning Systems: a
0272684X211022158. Literature Review, based on IEEE Explore Digital Library. In: 2020 IEEE Global
Naidu S. Instructional designs for optimal learning. Handbook of distance education Engineering Education Conference (EDUCON) IEEE; 2020. p. 284–289.
2007;p. 247–258. Siemens, G., 2008. Learning and knowing in networks: Changing roles for educators and
Nikou, S., Maslov, I., 2021. An analysis of students’ perspectives on e-learning designers. ITFORUM Discuss. 27 (1), 1–26.
participation–the case of COVID-19 pandemic. Int. J. Inf. Learn. Technol. Silva, C., Fonseca, J., 2017. Educational data mining: a literature review. Eur. MENA
Nisbet, R., Elder, J., Miner, G., 2009. Handbook of Statistical Analysis and Data Mining Coop. Adv. Inf. - tion Commun. Technol. 87–94.
Applications. Academic press,. Silva, D., Randombage, P., Gankanda, W., Abeysekara, N., Edirisinghe, V.,
Ong, C.S., Lai, J.Y., Wang, Y.S., 2004. Factors affecting engineers’ acceptance of Chandrathilake, M., 2021. Evaluation of web-based distance education of medicine
asynchronous e-learning systems in high-tech companies. Inf. Manag. 41 (6), for undergraduates in Sri Lanka. Sri Lanka J. Obstet. Gynaecol. 43 (1).
795–804. Soni, V.D., 2019. IOT connected with e-learning. Vishal Dineshkumar Soni. IOT Connect.
Pandit, D., Agrawal, S., 2021. Exploring challenges of online education in COVID times. e-Learn. Int. J. Integr. Educ. 2 (5), 273–277, 2019.
FIIB Bus. Rev. p. 2319714520986254. Stoetzel, L., Shedrow, S., 2021. Making the Transition to Virtual Methods in the Literacy
Papamitsiou, Z.K., Economides, A.A., 2014. Learning analytics and educational data Classroom: Reframing Teacher Educa- tion Practices. In: Excelsior: Leadership in
mining in practice: a systematic literature review of empirical evidence. J. Educ. Teaching and Learning, 13, pp. 127–142.
Technol. Soc. 17 (4), 49–64. Su, Z., McDonnell, D., Wen, J., Kozak, M., Abbas, J., Šegalo, S., et al., 2021. Mental
Peña-Ayala, A., 2014. Educational data mining: A survey and a data mining-based health consequences of COVID-19 media coverage: the need for effective crisis
analysis of recent works. Expert Syst. Appl. 41 (4), 1432–1462. communication practices. Glob. Health 17 (1), 1–8.
Prabha, S.L., Shanavas, A.M., 2014. Educational data mining applications. Oper. Res. Su, Z., Cheshmehzangi, A., Bentley, B.L., McDonnell, D., Šegalo, S., Ahmad, J., et al.,
Appl.: Int. J. (ORAJ) 1 (1), 23–29. 2022. Technology-based interventions for health challenges older women face amid
Priatna, T., Maylawati, D., Sugilar, H., Ramdhani, M., 2020. Key success factors of e- COVID-19: a systematic review protocol. Syst. Rev. 11 (1), 1–8.
learning implementation in higher education. Int. J. Emerg. Technol. Learn. (iJET) Sumathi, S., Sivanandam, S., 2006. Introduction to Data Mining and its Applications, vol.
15 (17), 101–114. 29. Springer.
Qiao, P., Zhu, X., Guo, Y., Sun, Y., Qin, C., 2021. The development and adoption of online Sun, P.C., Tsai, R.J., Finger, G., Chen, Y.Y., Yeh, D., 2008. What drives a successful e-
learning in pre-and post-covid-19: combination of technological system evolution Learning? an empirical investigation of the critical factors influencing learner
theory and unified theory of acceptance and use of technology. J. Risk Financ. satisfaction. Comput. Educ. 50 (4), 1183–1202.
Manag. 14 (4), 162. Swan, K., 2020. Teaching and learning in post-industrial distance education. An
Qureshi, R., AlManna, Y.M., Deshmukh, A.P., 2013. Big data: growing pressure on global Introduction to Distance Education Rout- ledge, pp. 67–89.
storage by data created on social networking sites. Int. J. Comput. Sci. Manag. Res. 2 Tang, H., Xing, W., Pei, B., 2019. Time really matters: Understanding the temporal
(3), 1876–1881. dimension of online learning using educational data mining. J. Educ. Comput. Res.
Radha, R., Mahalakshmi, K., Kumar, V.S., Saravanakumar, A., 2020. E-Learning during 57 (5), 1326–1347.
lockdown of Covid-19 pandemic: a global perspective. Int. J. Control Autom. 13 (4), Tawafak R., AlFarsi G., Jabbar J., Malik S.I., Mathew R., AlSidiri A., et al. Impact of
1088–1099. Technologies During COVID-19 Pandemic for Improving Behavior Intention to Use
Rahmat, T.E., Raza, S., Zahid, H., Abbas, J., Sobri, F.A.M., Sidiki, S.N., 2022. Nexus E-learning 2021;.
between integrating technology readiness 2.0 index and students’e-library services Thai-Nghe, N., Drumond, L., Krohn-Grimberghe, A., Schmidt-Thieme, L., 2010.
adoption amid the COVID-19 challenges: implications based on the theory of Recommender system for predicting student performance. Procedia Comput. Sci. 1
planned behavior. J. Educ. Health Promot. 11. (2), 2811–2819.
Rajab M.H., Gazal A.M., Alkattan K. Challenges to online medical education during the Tomasevic, N., Gvozdenovic, N., Vranes, S., 2020. An overview and comparison of
COVID-19 pandemic. Cureus 2020;12(7). supervised data mining techniques for student exam performance prediction.
Rakic S., Pavlovic M., Softic S., Lalic B., Marjanovic U. An evaluation of student Comput. Educ. 143, 103676.
performance at e-learning platform. In: 2019 17th International Conference on Vai, M., Sosulski, K., 2011. Essentials of Online Course Design: A Standards-Based Guide.
Emerging eLearning Technologies and Applications (ICETA) IEEE; 2019. p. 681–686. Routledge.
Rakic, S., Tasic, N., Marjanovic, U., Softic, S., Lüftenegger, E., Turcin, I., 2020. Student Vialardi-Sacín C., Shaftr L., Braver J., Ortigosa A. Recommendation in higher education
performance on an E-learning platform: mixed method approach. Int. J. Emerg. using data mining techniques. Universidad de Cordoba; 2009.
Technol. Learn. 15, 2. Wan, S., Niu, Z., 2019. A hybrid e-learning recommendation approach based on learners’
Raval, K.M., 2012. Data mining techniques. Int. J. Adv. Res. Comput. Sci. Softw. Eng. 2 influence propagation. IEEE Transac- tions Knowl. Data Eng. 32 (5), 827–840.
(10). Wang, X.Y., Li, G., Malik, S., Anwar, A., 2021. Impact of COVID-19 on achieving the goal
Rawat, K.S., Malhan, I., 2019. A hybrid classification method based on machine learning of sustainable development: E-learning and educational productivity. Econ. Res.
classifiers to predict performance in educational data mining. Proceedings of 2nd ̀ on. Istraživanja 1–17.
-Ek
International Conference on Communication, Computing and Networking. Springer,, Wang, Y.S., 2003. Assessment of learner satisfaction with asynchronous electronic
pp. 677–684. learning systems. Inf. Manag. - Ment. 41 (1), 75–86.
Ray, S., Saeed, M., 2018. Applications of educational data mining and learning analytics Whitley L.A. Educational data mining and its uses to predict the most prosperous
tools in handling big data in higher education. Appl. big data Anal. Springe 135–160. learning environment 2018.
Rodrigues, H., Almeida, F., Figueiredo, V., Lopes, S.L., 2019. Tracking e-learning through Wiliam, D., Thompson, M., 2008. Integrating Assessment With Learning: What Will it
published papers: a systematic review. Comput. Educ. 136, 87–98. Take to Make It Work? Routledge.
Rodrigues, M.W., Isotani, S., Zarate, L.E., 2018. Educational Data Mining: a review of Witten, I., Frank, E., Hall, M., Pal, C., 2017. Data Mining: Practical Machine Learning
evaluation process in the e-learning. Telem. - atics Inform. 35 (6), 1701–1717. Tools and Techniques. Elsevier Inc.
Romero, C., Ventura, S., 2010. Educational data mining: a review of the state of the art. Wolfinger S. An exploratory case study of middle school student academic achievement
IEEE Trans. Syst., Man, Cybern., Part C. (Appl. Rev. ) 40 (6), 601–618. in a fully online virtual school. Drexel University; 2016.
16
Yao, J., Ziapour, A., Abbas, J., Toraji, R., NeJhaddadgar, N., 2022. Assessing puberty- Zhang, D., Zhao, J.L., Zhou, L., Nunamaker Jr, J.F., 2004. Can e-learning replace
related health needs among 10–15-year-old boys: a cross-sectional study approach. classroom learning? Commun. ACM 47 (5), 75–79.
Arch. De. Pédiatrie 29 (4), 307–311. Zhang W., Qin S. A brief analysis of the key technologies and applications of educational
Yildiz, E.P., Cengel, M., Alkan, A., 2020. Current trends in education technologies data mining on online learning platform. In: 2018 IEEE 3rd International Conference
research worldwide: Meta-analysis of studies between 2015-2020. World J. Educ. on Big Data Analysis (ICBDA) IEEE; 2018. p. 83–86.
Technol.: Curr. Issues 12 (3), 192–206. Zhou, Y., Draghici, A., Abbas, J., Mubeen, R., Boatca, M.E., Salam, M.A., 2022. Social
Yu, S., Abbas, J., Draghici, A., Negulescu, O.H., Ain, N.U., 2022. Social media application media efficacy in crisis management: effectiveness of non-pharmaceutical
as a new paradigm for business communication: the role of COVID-19 knowledge, interventions to manage COVID-19 challenges. Front. Psychiatry 1732.
social distancing, and preventive attitudes. Front. Psychol. 13. ZoriÄ, A.B., et al., 2020. Benefits of educational data mining. J. Int. Bus. Res. Mark. 6 (1),
Zarei, S., Mohammadi, S., 2021. Challenges of higher education related to e-learning in 12–16.
developing countries during COVID-19 spread: a review of the perspectives of
students, instructors, policymakers, and ICT experts. Environ. Sci. Pollut. Res. 1–7.
17

Aulakh Et Al - 2023 - E-Learning Enhancement Through Educational Data Mining With Covid-19 Outbreak

Uploaded by

Copyright:

Available Formats

Aulakh Et Al - 2023 - E-Learning Enhancement Through Educational Data Mining With Covid-19 Outbreak

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aulakh Et Al - 2023 - E-Learning Enhancement Through Educational Data Mining With Covid-19 Outbreak

Uploaded by

Copyright:

Available Formats

Since January 2020 Elsevier has created a COVID-19 resource centre with

free information in English and Mandarin on the novel coronavirus COVID-

Elsevier hereby grants permission to make all its COVID-19-related

Contents lists available at ScienceDirect

International Journal of Educational Development

E-learning enhancement through educational data mining with Covid-19

Fig. 1. Overview of the abstract.

Fig. 3. ’Extraction of Knowledge’ component of EDM.

Fig. 4. Educational systems which contribute data for analysis.

Fig. 5. Types of data used for Educational Data Mining.

Table 1 groups (Sangrà et al., 2012):

Fig. 6. Recent E-learning trends.

Fig. 7. Key areas of Educational Data Mining application in E-learning.

Table 3 Table 3 (continued )

systems in developed countries gained ground in 2020 and 2021, as is Table 5

Table 6 from learner data, which is typically obtained by an intelligent

interactive e-learning methods has unfolded numerous opportunities for

7.2. Future research directions Author statement

You might also like