=Paper=
{{Paper
|id=Vol-2280/paper-28
|storemode=property
|title=Enhancing Moodle to Adapt to Students Different Learning Styles
|pdfUrl=https://ceur-ws.org/Vol-2280/paper-28.pdf
|volume=Vol-2280
|authors=Loreta Leka,Alda Kika
|dblpUrl=https://dblp.org/rec/conf/rtacsit/LekaK18
}}
==Enhancing Moodle to Adapt to Students Different Learning Styles==
https://ceur-ws.org/Vol-2280/paper-28.pdf
Enhancing Moodle to adapt to students different
learning styles
Loreta Leka Alda Kika
Informatics Department Informatics Department
Faculty of Natural Sciences Faculty of Natural Sciences
University of Tirana University of Tirana
loreta.leka@fshn.edu.al alda.kika@fshn.edu.al
A very well known LMS is Moodle (Modular Object
Oriented Developmental Learning Environment). It is
Abstract very popular because it is open source and also can
run on different platforms. In Moodle it is also pos-
The e-learning management systems are gain- sible to add modules[LGH14]. Despite these advan-
ing popularity in the academic community tages, Moodle also has some limitations. The same
offering many benefits for better and eas- content is offered to all the students despite their dif-
ier learning. Although there exist many e- ferences. It does not have the ability to adapt to their
learning management systems almost all of knowledge level, preference of learning styles, cogni-
them do not offer adaptivity features that tive ability, goals etc. Learning styles are considered
would made the system adapt to the need and to be an important aspect that affects the process of
learning styles of the students. This paper fo- learning. Therefore, adapting to them could also make
cuses on the need to enhance an existing open this process more efficient. There are various attempts
source e-learning management system, Moo- done in order to build an adaptive e-learning system
dle, by building an adaptive module to meet based on learning styles. Some have built new systems,
students different learning styles. The aim is whereas some have used a LMS, and extended them to
to create an adaptive module and integrate it be adaptive. LMSs provide a great variety of features
to Moodle system. In this paper we review which can be included in the courses such as quizzes,
the most relevent studies related to this sub- forums, chats, assignments, wikis, and so on. As such,
ject, analyze the techniques they have used they have become very successful and are commonly
to detect learning styles and provide different used by educational institutions, but they provide very
content. Based on these results and our previ- little or, in most cases, no adaptivity[Gra07]. In this
ous work, we propose the architecture for this paper we are going to make a review of various tech-
adaptation module. niques used for building adaptive systems based on
learning styles , and propose to build a new module
1 Introduction for Moodle system, that will be able to identify stu-
Learning management systems(LMS) are very useful dents learning styles and deliver content based on their
to every type of corporation or education institution style. This system will be used and tested with stu-
whose aim is to help the employees or students to learn dents at our department with the aim to analyse the
better and easier. E-learning can be defined as learn- result and improve in the future. The rest of the pa-
ing using electronic means. With the massive use of per is structured in the following way. In the second
internet, e-learning management systems are imple- section some of the most relevant studies and technolo-
mented by using modern technologies for the construc- gies that are used are analysed. In the third section
tion of web systems. The students like to learn using the architecture of the adaptation module of moodle
e-learning management systems because of many ben- will be presented. In the last section we summarize
efits that it offers like the possibility to access the in- our results and future work.
formation anyplace and anytime. There are different
popular LMS used nowadays even in education system.
2 Related work in a specific way. The idea of the literature-based ap-
proach is to use the behavior of students in order to get
Extending a LMS to be adaptive can be beneficial, hints about their learning style preferences and then
because it will provide users with all the functionali- apply a simple rule-based method to calculate learn-
ties that LMS-s have, and also add the adaptive func- ing styles from the number of matching hints [Gra07].
tionality. This would require less time, by avoiding From the literature, we can see that data-driven ap-
programming the whole system and focusing on the proaches are more widely used. One reason can be that
adaptive module. To provide adaptivity with respect it is more familiar to computer science researchers be-
to learning styles, the learning styles of the learners cause they require gathering relevant information for
need to be first known by the system. In this section the user model and then use an AI classification algo-
we are going to make a general review of some adaptive rithm to automatically detect the learning style pref-
e-learning systems from the literature and the tech- erences. Whereas the literature based approach re-
niques they have used to detect learning styles.There quires having some knowledge of psychology and cog-
are two main approaches used for detecting learning nitive science to correctly estimate the importance of
styles, based on the literature : explicit and implicit. the hints [JAA14]. Some of the artificial intelligence
In the first one, instructors provide students with a techniques used in data driven approach are: Bayesian
questionnaire that will calculate their learning style. networks, Decision trees, neural networks etc. Below,
There are many questionnaires available by the learn- there are a few adaptive systems and the technique
ing style model creators. One of them is the ILS(Index they use. Cha et al. (2006) derived relevant patterns
of Learning Styles) questionnaire based on the Felder- for detecting learning styles from the literature and
Silverman model. For example, this explicit method then used Decision Trees and Hidden Markov Mod-
is used in [Sur14], where students have to fill out the els to learn the parameters of the model from data
questionnaires when they accessed for the first time about the behaviour of students and from reference
the adaptive course. After determining their learn- data including the learning style preferences identi-
ing style, the system adapts the material according fied by the ILS questionnaire[Gra07]. Another ap-
to that. CS383 (Carver, Howard, and Lane, 1999) is proach for automatic student modelling was investi-
another system that uses this technique. It incorpo- gated by Garca et al. (Garca et al., 2005, 2007). They
rates Felder-Silverman Learning Style Model(FSLSM). observed the behaviour of learners during an online
The system provides adaptivity based on the sens- course in the system SAVER and used Bayesian net-
ing/intuitive, visual/verbal, and sequential/global di- works for identifying learning styles based on the be-
mensions of FSLSM. The learning styles of the stu- haviour of students[Gra07]. Sometimes, a mixed ap-
dents are calculated based on their answers and stored proach is used. An explicit model is used for the ini-
in the student model. The adaptation is done at the tialization of user model, and then it is updated using
presentation level by showing different types of me- an implicit model. For example, this technique is used
dia elements according to learning styles[Gra07]. This in TANGOW system. Adaptation in this system is
technique has some disadvantages, students learning based on two dimensions of FSLSM: sensing/intuitive
style can change during time and the system does and sequential/global. After initializing the student
not adapt at real time, because the questionnaire is model by filling the ILS questionnaire, it is automati-
only in the beginning. Moreover, filling a question- cally updated by observing the learners actions in the
naire requires an additional amount of work from the course. When learners behave contrary to the deter-
students, and they may not be interested. Sometimes mined learning style, preference stored in the student
their answers can be arbitrarily and can mislead the model is updated [Gra07]. In a study by (Graf, 2007)
adaptation process. In the second approach the sys- , Moodle is used as a LMS to extent with the capa-
tem does the automatic detection of learning styles. bility of identifying student learning style. A mixed
It analyses students behavior from the interactivity model is also used here. After the user model ini-
with the system, without the need of any question- tialization based on ILS questionnaire, the model is
naire. Automatic detection techniques can be divided updated based on the analysis of user data. The time
into two subcategories: data-driven and literature- spent in the system, the number of logins and the num-
based [Gra07]. In the first one, a classifier is built ber of visited learning requests for additional learning
that classifies students in one dimension of their learn- objects were recorded and analyzed[MAl16].For this
ing styles. This is done by applying a data mining analysis two approaches were used: a Bayesian Net-
algorithm on data from students activity and build work approach as well as a rule-based approach. Since
student models. In the second approach a literature the accuracy of the results was better in the latter
approach is used. Based on the literature, learners case, the rule-based approach was implemented into a
with a preference for a specific learning style behave tool called DeLeS, which can be used to identify the
learning style of the students in any LMS [Elv09]. An- fill in the questionnaire in the beginning. This will
other approach is used in a study in [Des+12].In this help in building their initial user model. We will pro-
study, for creating an adaptive system based on Moo- vide adaptive content for the users with the identified
dle, a data mining clustering technique is used. After learning style and provide a default content for the bal-
collecting initial data based on a questionnaire and an anced one. This user model will be updated on regular
introductory course, they apply the clustering algo- basis. We will decide on the technique of the model
rithm to divide students into some groups, based on adaptation in the future, after studying and testing
their learning style. This technique is more efficient es- them with the initial dataset. The time of adaptation
pecially for courses with large group of students, where could be a configurable parameter in the system. The
adapting courses for a group of student is less difficult system will be able to identify students learning styles
and also less time consuming for teachers, who need to by exploring their activity using system logs, such as
adapt learning materials for group of students instead time spent in an activity or lesson, the frequency of
to each student [Des+12]. accessing certain type of materials, how much they
like to use forums, chats etc. According to learning
styles, the system should be able to provide different
3 Proposed Solution content.In 1an overview of the proposed systems ar-
As we can see from the literature, FSLSM is the chitecture is presented. Student Model is generated
most often used learning style model in these adaptive from the ILS Questionnaire as well as the Dynamic
systems, where some systems incorporate the whole Student Modeling Module. Activities Data represents
model and some systems include only some dimensions the module that will collect students data from their
of FSLSM. We aim to build an adaptive system using activity. The Adaptive Content Presentation module
adaptation based on FSLSM learning styles. We have will be responsible for choosing the appropriate mate-
also chosen to build this adaptive module, as an ex- rials from the Learning Objects Module according to
tension to Moodle LMS. Moodle can be seen as the the detected learning style. Instructors will add mate-
best LMS concerning adaptation issues. In a study rials and annotate them according to learning styles.
in [SB05]nine platforms were analysed in detail, re-
garding their adaptability, personalization, extensibil-
ity and adaptivity. Moodle obtained the best results
in the general as well as in the adaptation evalua-
tion. Moodle provides an adaptive feature called lesson
where learners can be routed automatically through
pages depending on their answer to a question af-
ter each page. Furthermore, the extensibility is sup-
ported very well by a documented API, detailed guide-
lines, and templates for programming. Being an open
source product, Moodle also enables third-party en-
hancement of its functionality by the addition of mod-
ules. In[LA18] we have investigated students learning
style in a group of students at our department based
on the ILS questionnaire which uses Felder-Silverman
learning style model. This study revealed that a con-
siderable amount of students were balanced on the
learning styles dimensions. One of the reasons for
that, can be that students do not have stable learning
styles and they can change during time or depending
from the course. Sometimes they can also give arbi-
trary answers which lead to wrong calculations. We
came to the conclusion that relying only on the ques-
tionnaires to identify learning styles is not a good ap-
proach. On the other hand, using only an automatic
approach would take more time and would also need a
lot of initial data in order to be accurate. We therefore
suggest to build an adaptive system based on learning Figure 1: System architecture
styles using a mixed approach. We will enhance Moo-
dle, and build a new module for this. Students will
4 Conclusion and future work [JAA14] Feldman Juan, Monteserin Ariel, and
Amandi Analı́a. “Automatic detection of
In this paper we proposed to build a module for mak-
learning styles: state of the art”. In: Artifi-
ing Moodle an adaptive e-learning system based on
cial Intelligence Review (2014).
learning styles. We investigated some of the most com-
mon techniques used in similar systems, from the liter- [LGH14] Prabhani Pitigala Liyanage, Lasith Gu-
ature. There are two main techniques used for build- nawardena, and Masahito Hirakawa. “Us-
ing the student model: explicit and implicit. Explicit ing Learning Styles to Enhance Learning
model is based only on a questionnaire, whereas the Management Systems”. In: International
implicit method is based on automatic learning style Journal on Advances in ICT for Emerging
detection. In the implicit model, two main techniques Regions (2014).
were used: data driven model and literature based [Sur14] Herman Dwi Surjono. “The Evaluation
model. The first one includes the use of data min- of a Moodle Based Adaptive e-Learning
ing algorithms to analyze students data, and classify System”. In: International Journal of In-
students to different student models. The second one formation and Education Technology 4
uses a simple rule based method, based on students (2014).
activity. In the first model, data mining algorithms
[MAl16] Samar M.Alkhuraiji. Dynamic adaptive e-
as Bayesian Networks, Decision Trees, Clustering al-
learning mechanism Based on learning
gorithm, etc. are used. We analysed some of the sys-
style. A thesis submitted to the University
tems and the technique they use. Based on the studies
of Manchester, 2016.
that we have analysed and Moodle capabilities we de-
fined and presented the architecture for our adaptation [LA18] Leka Loreta and Kika Alda. “A study
module. We will integrate the questionnaire in Moodle on student’s learning styles using Felder-
system, and decide about the algorithm that we will Silverman Model”. In: KNOWLEDGE, In-
use to automatically detect students learning styles. ternational Journal Budva, Montenegro
As mentioned in this paper, different solutions use dif- 23.1 (2018).
ferent data mining algorithms for this reason. We will
make an analyses about their performance and choose
one to implement. In this paper, we have proposed
that the adaptation will be done at specific intervals
of time. After the implementation, we aim to test this
technique and better analyze this solution.
References
[SB05] Graf Sabine and List Beate. “An Evalua-
tion of Open Source E-Learning Platforms
Stressing Adaptation Issues”. In: Advanced
Learning Technologies (2005).
[Gra07] Graf. Adaptivity in Learning Management
Systems Focussing on Learning Styles. Vi-
enna University of Technology, 2007.
[Elv09] Popescu Elvira. Diagnosing Students’s
Learning Style in an Educational Hyper-
media System. Cognitive, Emotional Pro-
cesses in Web-based Education: Integrat-
ing Human Factors, and Personalization.
Advances in Web-Based Learning Book Se-
ries, IGI Global, 2009.
[Des+12] Despotović-Zrakić et al. “Providing Adap-
tivity in Moodle LMS Courses”. In: Educa-
tional Technology Society 15 (2012).