IJIRST –International Journal for Innovative Research in Science & Technology| Volume 1 | Issue 10 | March 2015

ISSN (online): 2349-6010

A User-Intelligent Adaptive Learning Model For

Learning Management System Using Data Mining
And Artificial Intelligence
Mrs. Subitha Sivakumar [Link] Venkataraman
Head of Department Teaching Fellow
Department of Software Engineering Department of Faculty of Computing
Botho University, Gaborone, Botho University, Gaborone,
Botswana Botswana

Mr. Cross Gombiro

Sr. Lecturer / Module Leader
Department of Faculty of Computing
Botho University, Gaborone, Botswana

Abstract
Entire world is revolving towards digital space as a result of the Internet and other emerging web technologies are helping the
society to reach the universe. ICT and e-learning are growing radically fast and have captured a major role in higher educational
systems. To develop the coast-to-coast purposes, institutions are changing the teaching style from chalk and talk to Learning
Management System (LMS), called e-learning systems. Impact of e-learning introduces various delivery methods for teachers
and different platforms to learn for learners. Teaching through LMS is most challenging because the difference in the learning
styles and the nature of the course. Materials used for delivering the course are mostly static for all types of learners. Students are
learning from the LMS tools, but not most effectively. To overcome this issues, the author(s) are recommending a new user
intelligent adaptive learning model to be used in LMS. This allows the model to identify the learners learning style and endorses
the appropriate learning materials for the learners. This method of delivering will be effective and efficient for different LMS
learners.
Keywords: Learning Management System, Data Mining, Artificial Intelligences, e-learning, adaptive learning
_______________________________________________________________________________________________________

I. INTRODUCTION
Numerous Educational institutions have moved from traditional classroom delivery to Learning management Systems where web
technologies and cognitive theories have been embedded in E-Learning technologies. (Education, n.d.) shows e-learning as
―learning method and a technique for the presentation of academic /curricula via the Internet, intranet or any other electronic
media inclusive of multimedia, compact discs, satellites, or other new education technologies‖.
There are several e-learning tools that are accessible freely like Moodle and commercial like Black board etc. E-learning
supports organizations in many ways such as improving exercise outlays, decreases the equipment and environmental structures,
increasing the results, standardization and consistency delivery. E-learning will also support learners as to have real time
experiences, free from class as any time education, will recuperate the retention, improve the individual learning, will encourage
sharing and knowledge gaining. E-learning Technologies are having issues in delivering the materials for learners. Instructors are
framing or creating the materials for delivery and posting in the LMS platforms for learners to use the materials, but not thinking
how useful the materials are? And how learners are adapting to the teaching styles are helpful? In e-learning, the materials will
be used by different types of learners and some may be interested learning through audio, video, visual or presentations. But
normally the materials are standard for all learners.
The main purpose of the research is to come up with an innovative adaptive tool for e-learning platform that can be used as an
addin tool for Learning Management Systems. The tool can help with coming up with course contents that are tailor made to
leaner’s expectations, learning style and learners motivation.

II. BACKGROUND STUDY

According to (Yeu-Hui & Chiung-Wen, 2013) they describe the role that new technology play as ―New technologies are
changing rapidly and have the potential to play an important role in teaching and learning. Mobile and wireless technology
brings new opportunities to traditional classroom-based learning as well as to lifelong learning outside the classroom‖. Change of
teach from classroom to e-learning supports the learners to learn anytime, anywhere in saving the time, value and location.

All rights reserved by [Link] 78

 A User-Intelligent Adaptive Learning Model For Learning Management System Using Data Mining And Artificial Intelligence
(IJIRST/ Volume 1 / Issue 10 / 018)

(Docimini & Palumbo, 2015) mentions e-learning as ―a dynamic and immediate learning environment through the use of internet
to improve the quality of learning by providing students with access to resources and services, together with distant exchange
and collaboration‖. In another aspect, e-learning improves the quality structure of an organization and also the learner’s freedom
of learning.
Tutors or subject experts are developing the delivery content, activities and exercises for the modules as generic. Problem is
that an e-learning environment will have learners with different learning styles as well as different cultures from different states
and regions. Generic materials may not help in all the cases. (Doherty, 2006) explains ―The lack of effective interaction between
students and the instructor and among students in the course often results in unsatisfying learning experiences and/or students
dropping out or failing online courses‖. Learners will have challenges or may not be interested to use the materials and it can
forced to stop-reading. (Kosta & Boris , 2015) refers to ―to demonstrate the design and evaluation of an adaptive, intelligent and,
most important, an individualised intelligent tutoring system (ITS) based on the cognitive characteristics of the individual
learner‖.
Learning materials should be based on the learning style. Some learners may be interested to learn the materials through
video’s, audio, and animation, verbal or through text. The Author(s) are recommending a new User-Intelligent adaptive Learning
Model for LMS to improve the learner’s interest and involvement based on their learning style. To achieve this, the two main
techniques data mining and artificial intelligences have begun used for drilling to find the learning style and based on the model
will suggest the suitable learning materials.

III. OBJECTIVES
The main objectives of the research are listed below:
- To apply Data Mining Techniques to categorize learners learning styles
- To apply artificial intelligent techniques to come up with an enhanced new e-learning platform
- To explore contents necessary for learners based on the learning style through the new platform

IV. TOOLS AND METHODOLOGIES

A new User-Intelligent Adaptive Learning Model for Learning Management System is shaped for e-learning platform as shown
in Figure 1.

Fig. 1: User-Intelligent Adaptive Learning Model

A. The Overhead Presented Model Has Four Stages:

Under this section we can use present tense like : ―Students demographics data are gathered from the student profile which will be collected‖
 Stage 1: Students demographics data are gathered from the student profile which will be collected at the time of students
admission, and their performance data will be monitored and recorded in their profile. The web logs in the Learning
management system help us to get the information about student interaction in the LMS on a specific module. (Hanna,
2004) describes e-learning is a source which ―automatically collected data is web server logs; these are vast collections
of data relating the accessing of specific web pages ‖. According to (Zorrilla, et al., 2005) refers ―students typically log
in to such systems, keeping track of users and sessions—a major hurdle in examining server logs‖. For course
personalization and identifying the learning style of learners a questionnaire should be prepared to know the learner

All rights reserved by [Link] 79

 A User-Intelligent Adaptive Learning Model For Learning Management System Using Data Mining And Artificial Intelligence
(IJIRST/ Volume 1 / Issue 10 / 018)

preliminary knowledge on the course as well to know the learning characteristics. The student data are recorded from
above sources as the students activities and usage are used as a dataset for further processing in stage 2.
 Stage 2: The Student data set which was retrieved from Stage 1 will preprocessed using Feature selection technique to
clean and remove the redundant data as the data set gathered may have different types of unrelated data and this may be
difficult to characterize the student’s behavior so there is need to preprocess and find the most relevant attributes that
will be needed for mining. Now to classify the students based on their learning style we use Clustering algorithm.
(Chapple, 2015) defines cluster as ―technique used to place data elements into related groups without advance
knowledge of the group definitions‖. By using this, the students can be grouped based on their learning styles and
behavior.
 Stage 3: In this third stage, the students learning style are found based on the student data
 Stage 4: Felder and Silverman (1988) have defined four learning style dimensions, which can relate for a particular
learning style and behaviours. Felder and Silverman describe the learning style as Sensing-Intuitive, Active-Reflective,
Visual-Verbal and Sequential-Global learners as shown in the Figure 2.

Fig. 2: Felder Learning Style Dimensions

The clustered group of learner from stage 2 can be mapped within the four dimensions of learning styles. For example, if a
student learning style is identified as visual-verbal, then this model should provide the learning material based on pictures,
diagram, flowcharts, written and spoken explanations. Now the learning materials can be personalized based on the learning style
for the learners who are clustered as a group as discussed in stage 2.
Stage 4: The course specialist or course developer should create the content’s for all types of learning styles (i.e.,) the rich
learning contents should be prepared as video based content, audio based content, handout based activities, animated content,
pictures based brilliance and hypertext materials etc.,
The AI techniques will be used to facilitate the extraction of learners needs bases on variables that influence learners academic
performance. Artificial intelligent has always been a branch of computers that can be used to facilitate intelligent tutoring,
decision support, assessment of learning outcomes and theories of teaching in education. AI and associated applications under e-
learning play a crucial role in providing e-learning systems that are intelligence and in some instances behavior based.
Considering different cultures, learning styles, different teaching styles there is needed to incorporate intelligence in learning
management systems to enhance learners behaviour. Such techniques such as Fuzzy logic, incorporating intelligence, machine
learning and confidence factors in Knowledge Based Systems and Decision support systems can extensively benefit LMS where
deep searching techniques can be used to come up with content that highlights behaviour of a learner. Once the learners profile
deduce from Data mining an adaptive e-learning style can be adopted by embedding instructural design theory into the LMS. The
learners can use the Virtual LMS to access the recommended content of learning style by model. Because of this, the leaners will
have more interest in utilizing the system.

V. CHALLENGES
The innovative developed model has certain confines
- Instructors or subject expert need to develop different materials based on the learning style
- Time consuming and cost of development will increase
- Learners will be forced to use only the suggested materials

VI. CONCLUSION
The research is based on the usages and performance of e-learning tools in the educational fields and found the materials should
be characterized constructed on the learners learning style. The author(s) recommends to use this model to be effective and

All rights reserved by [Link] 80

 A User-Intelligent Adaptive Learning Model For Learning Management System Using Data Mining And Artificial Intelligence
(IJIRST/ Volume 1 / Issue 10 / 018)

efficient for e-learning tools like Learning Management Systems. This model will motivate the learners to handle the e-learning
tools very successfully and the tutors need to be familiar with the characteristics in preparing the learning materials.

REFERENCES
[1] Chapple, M., 2015. Clustering. [Online] Available at: [Link]
[Accessed 02 2015].
[2] Docimini, G. & Palumbo, F., 2015. How to build an e-learning product: Factors for student/customer satisfaction. Business Horizons, Volume 20, pp. 87-
96.
[3] Doherty, W., 2006. An analysis of multiple factors affecting retention in web-based community college courses, Internet High. Education, Volume 9, pp.
245-255.
[4] Education, M. o. H., n.d. E-Learning and Distance Education, Kingdom of Saudi Arabia: International Education and Conference of Higher Education.
[5] Hanna, M., 2004. Data mining in the e-learning domain. Campus-Wide Information Systems, Volume 21, pp. 29-34.
[6] Kosta , D. & Boris , A., 2015. TECH8 intelligent and adaptive e-learning system: Integration into Technology and Science classrooms in lower secondary
schools. ELSEVIER: Computers & Education, Volume 82, pp. 354-365.
[7] Yeu-Hui, C. & Chiung-Wen, T., 2013. Enhancing nursing students’ medication knowledge: The effect of learning materials delivered by short message
service. Computers & Education, Volume 61, pp. 168-175.
[8] Zorrilla, M. et al., 2005. Web Usage Mining Project for Improving Web-Based Learning Sites. Spain, 10th International Conference on Computer.

All rights reserved by [Link] 81