CC BY
28
Научни трудове на Съюза на учените в България-Пловдив. Серия В. Техника и технологии, естествен ии хуманитарни науки, том XVI., Съюз на учените сесия "Международна конференция на младите учени" 13-15 юни 2013. Scientific research of the Union of Scientists in Bulgaria-Plovdiv, series C. Natural Sciences and Humanities, Vol. XVI, ISSN 1311-9192, Union of Scientists, International Conference of Young Scientists, 13 - 15 June 2013, Plovdiv.
LEARNING STYLE DETERMINATION IN E-LEARNING SYSTEM
Aleksandar Kotevski(1), Cveta Martinovska (2) and Radmila Kotevska(3) (1): University "St.Kliment Ohridski", Faculty of Law -R.Macedonia,Bitola (2): University "Goce Delcev", Faculty of Computer Science, R.Macedonia,Stip
(1): University "St.Kliment Ohridski", Faculty of Technical Science
-R.Macedonia, Bitola
e-mail(1): [email protected] e-mail(2): [email protected] e-mail(3): [email protected]
Abstract
The goal of this paper is to make combination of VARK classification and David Kolb's model to detect the student learning style and using the results in the e-learning process. Furthermore, system is going to update default learning style based on user behaviour and its responses. The main idea is that the process of e-learning will be more effective if students receive learning material in format that is adequate to their preferred learning style. On the other words, the system should meet the needs of students, showing learning materials in acceptable format and style to the user.
Teachers will post learning materials in several different forms. Then, based on student learning style, system is going to delivery learning content to the students.
Introduction
Today, there are a large number of e-learning systems that are using in the process of education in high school all over the world. Some of them have ability for adaptation to the student need and their knowledge level, goal, learning style and so on. The common characteristic is that they aim to improve the quality of their. In this context, they deliver the most adequate content to users, based on their requirements and learning styles. E-learning system has a large number of learning materials, posted in different format and style that is predefine by system administrator or by some templates. Main problem is that students have different learning style - some of them prefer to listen and talk other to using visual medium. Some of them want to analyze a text, other to learn through examples and real problem explanation. That's why it's very important to deliver the learning materials based on student learning style. Additional, it will enable the learner to improve the effectiveness of its approach to learning and to exploit its own resources
[2]. Otherwise, delivering learning materials that are not adequate to the student learning style will produce no productivity and more time consumption while student using the learning materials.
what is learning style?
Individual learning styles differ, and these individual differences become even more important in the area of education [4]. It is known that we all have different approaches to learning. Psychologists call these individual differences learning styles. Learning styles consist of a combination of motivation, engagement, and cognitive processing habits, which then influence the use of metacognitve skills such as situation analysis, self-pacing, and self-evaluation to produce a learning outcome [1]. They are based on the research results of cognitive psychology about processing information, active learning and the structure of information. The learners prefer intuitively some forms of information and a specific way of action over others when reaching quality learning. The division of learning styles is based on that. (Vainionpaa 2006). Learning styles refer to how individuals prefer to organize and represent information (Reed and Oughton 1997). As Coffield (2004) says the knowledge of learning styles can be used to increase students' self-awareness and metacognition of their strengths and weaknesses as learners. According to Jantan and Razali (2002), psychologically, learning style is the way the student concentrate, and their method in processing and obtaining information, knowledge, or experience. On the other hand, from the cognitive aspect, learning style can be referred to various methods in perception creation and information processing to form concepts and principles (Fleming & Baume 2006) [4].
The most popular technique for learning detection are David Kolb's model, Peter Honey and Alan Mumford's model, Anthony Gregorc's model, Sudbury model of democratic education, Neil Fleming's VAK/VARK model and so on. Each of them has own way to determinate the most adequate learning style of learners. Our proposed recommended agent is going to use combination of VARK classification and David Kolb's model to detect the student learning style.
David Kolb's model
The David Kolb's model is based on the Experiential Learning Theory, as explained in his book Experiential Learning: Experience as the source of learning and development [5]. Kolb's learning theory includes four different learning styles, which are based on a four-stage learning cycle. The learning cycle stages are:
1.Converger - abstract conceptualization and active experimentation, they making practical applications of ideas and using deductive reasoning to solve problems
2.Diverger - experience and reflective observation, imaginative and are good at coming up with ideas and seeing things from different perspectives
3.Assimilator - abstract conceptualization and reflective observation, they are creating theoretical models by means of inductive reasoning
4.Accommodator - concrete experience and active experimentation, engaging with the world and actually doing things instead of merely reading about and studying them [6]
Users need to complete the questionnaire online or on paper. In our case, they are going to complete online.
VARK classification
The VARK is model for detecting learning styles by providing questionnaires with 16 questions. It's authored by Fleming and Mills and has been used as a guide to help people learn more effectively. VARK is stands for Visual, Aural, Read/Write, Kinesthetic. It is VAK modification and includes a systematic presentation of questions to identify preferences for the ways information and ideas can be taken in or put out. The VARK model is based on principles of sensory perception so the instructional methods must be a stimulus for the student to gain any understanding of the subject [7].
Users need to complete the questionnaire online o r on paper. In our case, they are going to compete online. They can have more than one answer per (question, so they get a profile of four scores - one for each modaHty.
Proposed system
The go al of this paper is adaptation of learning materials Imed on students learning style. It means that the system needs to determinate themost aoequate learning style first. The student teaming style is going to be dehected via VARK and DaviuKotb's model. Tampers will post learning; materials as learning obj ects with different elemen(s (text, audio, Rideo, real examples, rimuration, presentation). System is going to select and present the adequate parts from learning object l^sen on studont learning style.
Eac h student has ofofile which cantains Serormation about the undent. Two learning styles are saved in the student profile bo tinee parameters:nrIRe, weight and author.
For- each stndent, system beeds to sdect one oa them as default learning style.
The default learnibg style can be selected in two manners:
- From htndent beУaviout
- From history of using the system
System architecture
The general architecture of the proposed system contains two main software units:
- Student unit
- Teac hers unit
Student need to be registered to b e able to use learningmqterials (objects) that are post from Drtehchrrs. !n the process of" registering, student comptered two questiomiaires, define by VAgfK and DeoidKeIb's model for lfarning siyle detection. The resuh of preview staee is proposing two teaming styles: one from VAeRK claasification, ohes from David Kolb'! model. Both we ill be store in student profile and are useful parameters for the system when learning materials are delivering.
Figure 1: System architecture
How system works
After sign up process, two learning styles are stored in the student model. Next issue is how system will know which is default learning style - learning style that should be primary for delivery learning materials. Othe r style will be second option if student want to switch the style and format of delivering.
The default learning style can be selected in two manners:
- From student behaviouT
- From ^story of using the system
A) Detection default learning style based on student behaviour
1) System send teaming material in two formats (first one proposed from VARK classification, second one proposed from Kolb model)
2) °tudent gdt screen notification and dstailt about presentation of learning materials
3) Student selects one of them and learning materials are visiWe for the student
4) If dtyle ou distribution is acceptable for the student, he oonfrsms thiat. It initialize increment of weight field of the solocted learning style in student profile
5) Otherwise, student switches other screen. Id initialize oecrement of weight field value of the preview ieaпiing style m student profile and decrement weight field value of the selected learning atyle
Figure 2: Detection default learning style based on seudent behaviour
B) Detection default learning style based on history of using the system
1) System^d the learning material
2) System get irfotmalirn about autnor oa loaded material (hi s unique number)
3) System check in database to select all learning material from statistic view table from author (2) and student unique number is equivalent to logged student
4) It select rows from (3) and calculate total weight value for the list, approximately for both learning style
5) The learning style that has higher value is set as a default learning style Conclusion
Important aspect for system efficiency is its intelligent - ability to adapt the system based on user needs and habits. It's true mat each student has different learning style. Some of them preferred learning from audio materials, some of them learning from examples and practical situation. Others love to only to read text and so on. To be e-learning system more efficient, it has to be adaptive to student learning style. If system deliver learning material in inadequate form, then the materials are not very useful for students. Students will spend more time to understand the materials. That is opposite of e-learning system goal - to help to users and make learning process more easy and interesting.
That's why, good idea is using recommendation agent in e-learning system, which will delivery learning materials in the most adequate style, based on users need - his favorite learning style. Based on our own experience, it is not very particle if learning materials are deliver to students in format and style that is not adequate to them. That way of material delivering has negative implication to the learning process. Furthermore, materials that are deliver in inadequate learning style looks unusable and confusable for the students. That's why adaptation of learning materials is very critical aspect of each e-learning system.
In this paper, we are using two famous methods for learning style detection, VARK and David Kolb's model. System generate two learning style for each student, and with the two mention techniques system set one of them as default style for delivering learning materials.
References
[1] Ching-Chun Shih, Julia Gamon, Web-based learning: relationships among student motivation, attitude, learning styles, and achievement
[2] Essaid El Bachari, El Hassan Abdelwahed, Mohamed El Adnani, Design of an adaptive e-learning model based on learner's personality, Ubiquitous Computing and Communication Journal
[3] Owen Conlan, Cord Hockemeyer, Vincent Wade, Dietrich Albert, Metadata Driven Approaches to Facilitate Adaptivity in Personalized eLearning Systems
[4] K - Naser-Nick Manochehr, The Influence of Learning Styles on Learners in E-Learning Environments, Information Systems Department, Qatar University
[5] Kolb, David (1984). Experiential learning: Experience as the source of learning and development. Englewood Cliffs, NJ: Prentice-Hall. ISBN 0-13-295261-0.
[6] Smith, M. K. (2001). David A. Kolb on experiential learning. Retrieved October 17, 2008, from:http://www.infed.org/biblio/b-explrn.htm
[7] J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68-73.
[8] Sunita B Aher, Lobo L.M.R.J., A Framework for Recommendation of courses in E-learning System, International Journal of Computer Applications (0975 - 8887) Volume 35- No.4, December 2011
[9] Bourbia Riad, Seridi Ali, Hadjeris Mourad, and Seridi Hamid, An Adaptive Learning Based on Ant Colony and Collaborative Filtering, Proceedings of the World Congress on
Engineering 2012 Vol II WCE 2012, July 4 - 6, 2012, London, U.K.
[10] Lamia Berkani, Omar Nouali, Azeddine Chikh, Recommendation-based Approach for Communities of Practice of E-leaming
[11] Yi-Chun Chang, Wen-Yarn Kao, CУiУ-Ping Chu, Chung-Hm Chiu, A learning style classification mechanism for e-leaming, Computers & Education 53 (2009) 273-285
[12] Maria Zajac, Using learning styles to personalize odtine learning
[13] Fathi Essalmi, Leila Jemni Ben Ayed, MoУaRed Jemni, Kinshuk b, Sabine Graf, A fully personalization strategy of E-leaming scenarios, Computers in Human Behavior 26 (2010) 581-591
[14] Ana Lidia Fradzodi Velazquez, Said Assa!, Using learning styles to edУadce an E-leamuing system
