Recommended system for e-learning based on
personality type and learning style
Mrs. Yogita Mane
HOD, Department of Information Technology
Universal College of Engineering
Mumbai, India
yogita.mane@universal.edu.in
Mr. Anuj Hegishte
Department of Information Technology
Universal College of Engineering
Mumbai, India
anuj.hegishte@universal.edu.in
Mr. Mihir Dedhia
Department of Information Technology
Universal College of Engineering
Mumbai, India
mihir.dedhia@universal.edu.in
Mr. Aditya Nair
Department of Information Technology
Universal College of Engineering
Mumbai, India
aditya.nair@universal.edu.in
Abstract- The current e-learning management
systems contain a huge pool of data gathered from
multiple sources but main challenge that these
systems are facing is the provision of relevant and
quality content to the users and reduce the time
wasted by the users to search this content. Also,
with differences in the learning capabilities not all
users can undertake the same learning track for
understanding a particular content. Personal
Learning Environment (PLE) is an e-learning
concept that allows users to manage their learning
environment both in terms of content and process.
However, significant problems with PLE
implementation in distance learning are excessive
information and difficulties in finding the suitable
learning content for learners. To overcome these
problems, an experimental study was conducted to
explore a learning content recommendation system
for learners. Like the users of e-commerce system,
some students may also feel overwhelmed by the
available choices of material contents offered by the
e- learning system in which, it does not always suit
to their learning style. This is important as some
experts in educational psychology suggest that
students need to learn by following their personal
learning style. So, we would recommend e- learning
material to user according to user personality type
and learning style.
Keywords- Personalized Learning Environment,
recommendation system, e-learning, Myers-Briggs,
Kolb’s
learning
style,
knowledge-based
recommendation.
when needed. But, the learning style of each student is
different, consequently, learning progression and
pattern varies among students. A learning style is a
characteristic of psychological and cognitive behaviour
of a user under a learning phase. Hence, a good elearning system is one which not only recommends
concept based on knowledge level but also recommends
the type of learning material which will help the student
to learn or acquire a skill in the best way.
E-Learning systems provide several types of
educational content and resources for learning like
video tutorials, blogs, articles, e-books etc. Discovering
and searching adequate learning materials and domainrelated content becomes one of the key challenges in
such E-Systems.
Learning Management Systems provides the users with
an environment that enables them to manage and search
small units of content for better interactive learning.
Each learning resource has its own characteristics and
features, when it comes to the way in which the data is
presented, the structure of the data or the format of the
content, etc. The main challenges in developing an elearning portal are developing an initial user profile and
continuously updating the profile to adapt to the user’s
change in preferences, interests, and needs.
Every student has their own personal learning style.
According to experts in this area, when students are not
performing as well as they could be, there must be
another unique way to teach them. It is especially
important when comes to choose a perfect learning style
so that the students can have a good motivation to learn.
Our research aims to help the students to find their
preferred learning style. We propose an e-learning
recommender system design based on a knowledgebased logic approach which can help the students to
select the best e-learning material matched to their
personality type and learning style.
I. INTRODUCTION
Every e-learning system is designed with a goal to help
the student to establish his/her goals and help them
II. PROBLEM STATEMENT
With the advent in technology and with the perpetual
increase in the strength of the students and the number
of departments in the educational institutions, it is
laborious to exchange the study materials between the
students and the faculties.
The main objective of the E-Learning is to help the
students get over the traditional methods of learning
and make them accustomed to the internet where the
notes for their respective subjects are easily available.
E-Learning is an inexpensive, efficient, and
comfortable way for students to easily access notes
and an easier alternative to study for exams. In our
project we try to give recommended e-learning
materials to user based on their personality type and
their learning style so they can save time in finding
resources.
III.
REVIEW OF LITERATURE
Nowadays, it is a quite common technology used in ecommerce system to assist users in retrieval of relevant
items. Despite being remarkably successful in ecommerce area, the implementation of recommending
a system for education especially e-learning is still
unexplored. The use of recommending a systems for elearning can be beneficial for both students and the
instructors, as well as for the institutions.
There are four recommendation approaches –
Collaborative
filtering
(CF),
Content
based
recommendation (CB), Hybrid recommendation system
and Knowledge-based recommendation system.
Content based (CB): - Content-Based recommendation
is based on identifying characteristics that are like those
a user has preferred in the past and make
recommendations accordingly.
Collaborative filtering (CF): - Collaborative filtering
recommendation is based on user behaviour or user
ratings of recommended items. It recommends items
liked by similar users and explores diverse content. By
accessing a learner profile, RS can access information
about age, country, previous learning activities,
educational background, etc. With the help of this
information, RS can find learners with similar learning
preferences and suggest learning materials accordingly.
The CF algorithm finds either prediction ratings or
recommends a list of top-N items.
Hybrid recommendation: - Hybrid RS is the
combination of CB and CF which combines
characteristics of both approaches through mergers of
individual predictions into one or adding content
information to collaborative model or by weighted
average of content and collaborative recommendations
or getting final recommendations based on the
combined rankings.
Table 1. shows the comparison of Implementation and
advantages and disadvantages of Collaborative filtering
(CF) and Content based (CB) recommendation system
and Hybrid recommendation system:
Table 1. Filtering Methods
Disadvantages of above-mentioned method
As in above discussed methods Collaborative filtering
(CF), Content based recommendation (CB), Hybrid
recommendation systems these have a problem of cold
start that is we need user’s data based on its profile to
identify his/her personality or data of user in Learning
Management System (LMS) like Moodle to identify
his/her learning style. So, as we do not have such data
available our research and proposed model can have
best solution to use Knowledge-based recommendation
system in projects.
Knowledge - based recommendation: - Some
recommendation systems used in e-learning are
Semantics or Knowledge-Based. They include ContextBased and Ontology-Based approaches. Systems are
knowledge-based and frame knowledge of content and
about stakeholders of the recommendation process
through ontology. For e-learning that means that such
systems map learner- relevant learning resources
through the exploitation of relational knowledge. We
can also recommend by identifying users' personality
type and learning style.
IV.
PROPOSED METHODOLOGY
To collect student data, we can make a website and
conduct a small personality test (MBTI Test) and user
will submit that test also can include test to identify
user learning style.
Fig 1. Activity Diagram
Identifying Student Personality
As user submitted the test, we would analyze it and
identify user’s personality type. To Identify Student
Personality, we are using here The Myers-Briggs Type
Indicator (MBTI) Myers-Briggs evaluates personality
types and preferences through four aspects of
personality:
 Extroversion (E) or Introversion (I)
 Sensing (S) or Intuition (N)
 Thinking (T) or Feeling (F)
 Judging (J) or Perceiving (P)
The functionality of the e-learning system is illustrated
by the use-case diagram shown below.
(AC) and active experimentation (AE). They follow each
other in a cycle.
•
•
•
•
Fig 2. Use Case Diagram
The combined sets of these different preferences
provide 16 different personality types and are typically
symbolized by four letters to represent a person's
movement on the four scales. For example, ESTP stands
for Extroversion, Sensing, Thinking, and Perceiving,
which show the four preferences of highest occurrence
for this person.
The MBTI assessment highlights the distinct nature of
each leaner's preference.
Diverging (concrete, reflective): Tend to be
innovative and imaginative in doing things.
Analyze concrete situations from different point of
views and adapts by observation not action. It is
feeling oriented.
Assimilating (abstract, reflective): Prefer to
observe, think, and integrate the whole picture.
Tend to reason inductively and to build up models
and theories.
Converging (abstract, active): Prefer practical
application and problem solver. Decision maker
and good handler to technical issues.
Accommodating (concrete, active): Tend to try
and error concept does not prefer thought and
reflection. Get adapted to changing situations.
Easy going with people.
In addition to the sixteen personality types, each
personality type has one dominant preference (dominant
process) which is used with the highest confidence. It
routes our personality and outlines our motives and
goals.
Fig 4. Kolb’s learning cycle
Table 2. Learning styles and corresponding e-learning
recommendation
Fig 3. Personality Types
Identifying Student Learning Style
Also identify learning style based on personality type
and test submitted by user. To Identify Student
Learning Style, we are using here Kolb's Experiential
Learning Kolb model was developed by Professor
David Kolb. Kolb divided the learning cycle to four
simple stages.
Kolb's four Stages are: Concrete experience (CE),
reflective observation (RO), abstract conceptualization
Analysis and Recommendation System: Now we
would analyze users' data and can get his personality
and learning style based on that our recommendation
system will recommend e-learning material to user.
Fig 5. Kolb’s learning style related to personality types
Recommendations of E-Learning Material to
Student
And user will get recommended e-learning material on
the website.
Technologies Used
Frontend
• HTML5
• CSS3
• JavaScript
• Bootstrap5
Backend
• MongoDB
• NodeJS
• Express
Also, unlike many other types of psychological
evaluations, your results are not compared against any
norms. Instead of looking at your score in comparison
to the results of other people, the goal of the instrument
is to simply offer further information about your own
unique personality.
When working in group situations in school or at work,
for example, recognizing your own strengths and
understanding the strengths of others can be immensely
helpful. When you are working toward completing a
project with other members of a group, you might
realize that certain members of the group are skilled
and talented at performing actions. By recognizing
these differences, the group can better assign tasks and
work together on achieving their goals.
REFERENCES
[1] M. S. Halawa, E. M. Ramzy Hamed and M. E.
[2]
[3]
[4]
V. RESULTS AND DISCUSSION
The Myers-Briggs Type Indicator test can provide a lot
of insight into your personality, which is why the
instrument has become so enormously popular. It can
also be helpful for an individual to identify the perfect
learning style suitable for the user. Even without taking
the formal questionnaire, you can probably
immediately recognize some of these tendencies in
yourself.
VI.
CONCLUSION
According to the Myers & Briggs Foundation, it is
important to remember that all types are equal and that
every type has value.
First, the MBTI is not really a "test." There are no right
or wrong answers and one type is not better than any
other type. The purpose of the indicator is not to
evaluate mental health or offer any type of diagnosis.
[5]
Shehab,
"Personalized
Elearning
recommendation model based on psychological
type and learning style models," 2015 IEEE
Seventh International Conference on Intelligent
Computing and Information Systems (ICICIS),
2015, pp. 578-584.
Kamal, A., Radhakrishnan, S. “Individual learning
preferences based on personality traits in an Elearning
scenario” 2019 Education and Information
Technologies. 24. 1-29.
K. Jetinai, "Rule-based reasoning for resource
recommendation in personalized e-learning," 2018
International Conference on Information and
Computer Technologies (ICICT), 2018, pp. 150154.
F. Hidayat, D. D. J. Suwawi and K. A.
Laksitowening,
"Learning
Content
Recommendations on Personalized Learning
Environment Using Collaborative Filtering
Method," 2020 8th International Conference on
Information and Communication Technology
(ICoICT), 2020, pp. 1-6.
R. Turnip, D. Nurjanah and D. S. Kusumo,
"Hybrid recommender system for learning
material using content-based filtering and
collaborative filtering with good learners' rating,"
2017 IEEE Conference on e-Learning,
e- Management and e-Services (IC3e), 2017, pp.
61-66.