Econf3-Bahrain University-April 6-8-2010
AACWELS: Automated “Adaptive Content”
Web Based E-learning System For Teaching AI
Prof. Dr. Abdul Hamid M. Ragab & Abubakr Salim Bajnaid
King AbdulAziz University
Jeddah, Saudi Arabia
ahm_ragab@yahoo.com, abaj1424@hotmail.com
Abstract
Nowadays, web-based education is reaching a large number of learners through the
internet. It poses a valuable advantage over traditional classroom teaching. Recently
however, many researchers agree on the fact that learning materials shouldn’t just
reflect the teacher’s style only, but they should also be designed to satisfy learning
needs for all kinds of students and all kind of learning styles. This process still raises,
some problems need to be solved, among which matching suitable teaching contents
with the student's learning style. In this paper, we propose a design and
implementation of an Automated Adaptive Content Web based E-Learning System
(AACWELS) for teaching AI subjects. The system uses an adaptive course content
taxonomy based on a modified Felder's questionnaire depending on the learning styles
properties used by Honey and Mumford. The proposed model tends to pursue
adaptation according to obtained user profile. The lesson content is tailored to
individual users, taking into consideration a specific learning style and subject matter
learning motivation. These guidelines are based on pedagogical strategy and
motivation factor with a strong psychological background. A demonstration of this
system is available at the site http://www.aicurriculum.org. System performance
results show that the system is scalable, and students are able to learn and to
efficiently improve their learning process with such methodology, hence improving
their learning gain.
1 Introduction
In this section we investigate several types of adaptive eLearning systems based on
learning styles. Then, we discuss the methodologies used to allocate course contents
using such systems, and compare these methods with our proposed system.
1.1 Adaptive systems based on Learning Styles
In order to provide adaptation features to e-learning environments, the systems itself
has to provide adaptive methods for the learning and teaching process. Several
educational hypermedia systems that adapt to learning styles have been developed [14]. We shall shortly review some of them in the following, grouped by the learning
style that they are trying to respond.
The field-dependent (FD) versus the field-independent is a popular learning style,
as in the “Adaptive Educational System based on Cognitive Styles”, AES-CS Error!
Reference source not found.; this division shows similarities (in features and
treatment) with the global-sequential dimension of the Felder-Silverman model, as in
“Learning Styles Adaptive System”, LSAS Error! Reference source not found. and
1
Econf3-Bahrain University-April 6-8-2010
CS388 Error! Reference source not found.. FD learners prefer experiences in a global
fashion, adhere to structures, learn material with social content best, attend best to
material relevant to own experience, require externally defined goals and
reinforcements, need organization, be more affected by criticism, use observational
approach for concept attainment (learn best by using examples). Adaptive hypermedia
(AH) systems respond to these needs by providing navigational support tools (e.g.,
concept map, graphic path indicator, advanced organizer) understanding the structure
of the knowledge domain. In terms of adaptive hypermedia, these techniques fall into
the category of adaptive navigation support. FI learners perceive analytically, make
specific concept distinctions, little overlap, prefer impersonal orientation, may need
explicit training in social skills, are interested in new concepts for their own sake,
have self-defined goals and reinforcement, can self-structure situations, are less
affected by criticism, use hypothesis-testing approach to attain concepts. AH systems
respond with learner control options, allowing arbitrary succession of course material.
Some system allow explicit learner directed switch between the two treatments.
Other systems implement strategies corresponding to the different sensory
preferences, as in ARTHUR Error! Reference source not found., iWeaver Error!
Reference source not found., MANIC Error! Reference source not found., CS388 Error!
Reference source not found.. The adaptive presentation in AH can switch automatically
between textual – (text, hypertext), auditory - (sounds, streaming audio), visual (streaming video, slideshows, hypermedia) and kinaesthetic preferences (animations,
simulations, puzzles).
Other learning styles, such as the sensing-intuitive dimension in the FelderSilverman scale Error! Reference source not found., are used in Tangow Error!
Reference source not found.. Sensing learners tend to like learning facts, intuitive
learners often prefer discovering possibilities and relationships. Sensors often like
solving problems by well-established methods and dislike complications and
surprises; intuitors like innovation and dislike repetition. Sensors are more likely than
intuitors to resent being tested on material that has not been explicitly covered in
class. Sensors tend to be patient with details and good at memorizing facts and doing
hands-on (laboratory) work; intuitors may be better at grasping new concepts and are
often more comfortable than sensors with abstractions and mathematical formulations.
Sensors tend to be more practical and careful than intuitors; intuitors tend to work
faster and to be more innovative than sensors. Sensors don't like courses that have no
apparent connection to the real world; intuitors don't like "plug-and-chug" courses
that involve a lot of memorization and routine calculations. To respond to these types
of preferences, teaching strategies involve selection of appropriate meta-data (such as
concept attributes of the form ‘example’, ‘activity’, ‘theory’, ‘exercise’) and
corresponding AH adaptive presentation.
Kolb’s learning styles are applied, e.g., in INSPIRE Error! Reference source not
found.. Kolb learning styles model is based on two lines of axis (continuums): to a
task - (preferring to do or watch), and to emotional response (preferring to think or
feel). The resulting learning styles are: activist (accommodator: doing and feeling
preferences, or concrete-active); diverger (diverger: watching and doing, or concretereflective); theorist (assimilator: watching and thinking, or abstract-reflective);
pragmatist (converger: thinking and doing, or abstract-active). The AH response to
these styles is to show different sequences of information, by using adaptive
navigational support techniques with link annotation.`
2
Econf3-Bahrain University-April 6-8-2010
The “Adaptive Hypermedia Architecture” AHA! [14], is developed to support an online course with some user guidance through conditional (extra) explanations and
conditional link hiding. Compared to other adaptive systems like Interbook [15],
KBS-Hyperbook [16] and many others AHA! excelled in the area of simplicity.
Many of the systems above assess learning styles via psychometric questionnaires,
and thus classify learners into stereotypical groups before the actual learning starts,
and no updates are possible. AH actually offers the tools to bypass this problem as we
shall see.
The main difference to our approach is that we offer an authoring tool that can
build different adaptive strategies corresponding to different learning styles, but is not
bound by it. There are no explicit limitations to what exactly the adaptive strategies
may represent, or to what subjects and concept structures they can be applied, as shall
be seen in the following section. Table 1 summarizes the above existing systems and
the learning styles they implement compared with our proposed leaning style based
system.
3
Econf3-Bahrain University-April 6-8-2010
Table 1 Learning styles incorporated into adaptive systems.
1.2 Adaptive Learning Systems and course content handling
A free web-course about Hypermedia Structures and Systems is implemented in the
Adaptive Hypermedia Architecture (AHA) [14]. The AHA can be used to generate
conditional text, to adapt the link structure by link removal, link hiding and link
annotation. Preprocessor commands in HTML pages are used by CGI-scripts to filter
content fragments of a page and thus enable content adaptation. The same
preprocessor technique is used for link adaptation. Some systems assume that all
users of the system have the same background. Thus it is a general assumption rather
than a background characteristic. In these approaches, content-level adaptation is only
done in the AHA system. However, AHA! does not confine the author to using a
particular learning style model. The user may specify preferred learning style by
selecting the appropriate type from a drop down list.
4
Econf3-Bahrain University-April 6-8-2010
A review of adaptive educational hypermedia systems showed that the systems
ELM-ART [23], INTERBOOK [15], and AHA [14], focus on similar aspects. The
ELM-ART system and its successors ELM-ART II [24] and INTERBOOK [15] are
some of the first adaptive hypermedia systems which are used in the WWW. The
KBS system is very similar to the approaches in INTERBOOK and ELM-ART II .
The adaptation features, like link adaptation, goals, page sequencing, are present in
each of the three systems. However, the implementation strategies to obtain these
functionalities vary. Comparing INTERBOOK with KBS, we see a difference in the
domain description. INTERBOOK uses a conceptual network of the domain, while
KBS uses both conceptual model and knowledge model. Therefore the introduction of
slots as used in INTERBOOK is not necessary in KBS, as assumptions about
prerequisite and outcome knowledge are based upon the knowledge model.
SIETTE [25] is an example of a Web-based adaptive testing system. The only kind of
learning material it possesses is questions. The only thing it can do is to generate an
adaptive sequence of questions to assess student's knowledge. Table 2 summarizes
these systems and comparison with our proposed one
ILESA[26]: is a Web-based Intelligent Learning Environment for the Simplex
Algorithm developed to help teachers in the task of teaching the Simplex Algorithm is
described. The domain of the system is algorithmic, and can be included into a special
class of instructional domains in which learning is strongly related to problem
resolution: exercise based domains. From a pedagogic point of view, the system can
be classified in the category of coached problem solving systems. In the development
of the system, several subtasks have been accomplished: the expert system to solve
linear programming problems, the problem generator, the student model, the
diagnosis module and the engine. ILESA also has a dynamic help system, and a webbased interface that is very adequate for the task of teaching. It is capable of
proposing problems to each student at the right level of difficulty for his current state
of knowledge, to evaluate the student's answer and to decide the next instructional
action.
Table 2 shows a comparison between the above common systems and our proposed
one. The Adaptive annotation in Table 2 means that the system uses visual cues (i.e.: icons,
fonts, colors) to show the type and the educational state of each link. Table3 shows a
Comparison between adaptation techniques implemented in these systems.
Table 2 A comparisons between common Adaptive systems and our proposed one.
5
Econf3-Bahrain University-April 6-8-2010
Table 3 A Comparison between adaptation techniques.
2 Learning Style Models
Learning style is a unique way to each student learns[30], most effectively. According
to cognitive psychology, learning styles could be briefly defined as personal manners
to perceive and process information, and how they interact and respond to educational
stimuli. A learning style is described as being: a description of the attitudes and
behavior which determine an individual's preferred way of learning. We define the
following four learning styles
1. Type 1 (concrete, reflective): A characteristic question of this learning type is
"Why?" Type 1 learners respond well to explanations of how course material
relates to their experience, their interests, and their future careers. To be
effective with Type 1 students, the instructor should function as a motivator.
2. Type 2 (abstract, reflective): A characteristic question of this learning type is
"What?" Type 2 learners respond to information presented in an organized,
logical fashion and benefit if they have time for reflection. To be effective, the
instructor should function as an expert.
6
Econf3-Bahrain University-April 6-8-2010
3. Type 3 (abstract, active): A characteristic question of this learning type is
"How?" Type 3 learners respond to having opportunities to work actively on
well-defined tasks and to learn by trial-and-error in an environment that allows
them to fail safely. To be effective, the instructor should function as a coach,
providing guided practice and feedback.
4. Type 4 (concrete, active). A characteristic question of this learning type is
"What if?" Type 4 learners like applying course material in new situations to
solve real problems. To be effective, the instructor should stay out of the way,
maximizing opportunities for the students to discover things for themselves.
Kolb’s experiential learning theory [27] offers a flexible, effective framework for
understanding learning and for devising teaching strategies to enhance learning. He
named four learning styles: the accommodator, the assimilator, the diverger, and the
converger. Felder’s [22] proposes a variation to Kolb’s position, establishing
relationships among styles and the ways information is dealt.
Honey and Mumford [20] developed a very popular Learning Styles Questionnaire,
which categorized people by their preferred learning styles into:
1- ACTIVISTS: who involve themselves fully and without bias in new learning
experiences? They are open-minded, not skeptical, they tend to act first and
consider the consequences afterwards…
2- REFLECTORS: who like to stand back and ponder experiences and observe
them from many different perspectives? They collect data and prefer to think
about it thoroughly before coming to any conclusions…
3- THEORISTS: who adapt and integrate observations into complex but logically
sound theories. They think problems through in a vertical, step by step, logical
way…
4- PRAGMATISTS: who are keen to try out ideas, theories and techniques to see
if they work in practice? They positively search out new ideas and take the
first opportunity to experiment with applications. They tend to be impatient
with ruminating and open-ended discussions.
Also, Honey and Mumford [20] point out that there is an association between the
learning cycle and learning styles as shown in Fig1. Felder and Silverman[11], also
proposed a model for learning styles that indicates a person’s predilections on five
continua: Sensory/Intuitive, Visual/Verbal, Inductive/Deductive, Active/Reflective,
and Sequential/Global. The model divides learning styles into five categories: Active
and Reflective Learners, Sensing and Intuitive Learners, Visual and Verbal Learners,
Inductive and Deductive Learners, and Sequential and Global Learners
3 Proposed Leaning Style Questionnaire (LSQ) For AI Students
The Learning Style Questionnaire profiles students by their preference for each
learning style [31], i.e. strong preference for Activist, moderate preference for
Theorist, and low for both Reflector and Pragmatist., Fig.1.
To learn effectively, you need to keep moving around this cycle:
7
Econf3-Bahrain University-April 6-8-2010
Fig.1 Leaning Cycle and Learning styles
1. Experiencing – doing something;
2. Reviewing – thinking about what has happened;
3. Concluding – drawing some conclusions;
4. Planning – deciding what to do in the future.
Completing each stage is important not just for itself, but because it improves learning
in the next stage. We used Felder's questionnaire [22] form after we have modified it
depending on the properties of each learning style according to Honey and Mumford
[20]. Which is short and specific, and it is suitable for the educational purpose that it has
been designed for. Table(4) illustrates a comparison between several well known methods
and our proposed one. And Fig.2 shows the proposed questionnaire.
Table 4 Comparisons between LSQs
8
Econf3-Bahrain University-April 6-8-2010
Fig. 2 the Questionnaire
4 Proposed Pre-test for AI Students
In the proposed system, each student has to do a pre-test, for answering the questions
of the Questionnaire shown in Fig.2. And Fig.3 shows Pre-test operations for
determining student's Learning Style. Then, the system analyzes results of this pre-test
and chooses the learning style which has the high credits that tell the student tends to
learn with this style. The system also displays results to students.
5 Questionnaire’s Assessment
Based on the Index of Learning Styles ( ILS) questionnaire, preferences for learning styles are
measured by values between +11 and -11, with steps of +/- 2, for each learning style
dimension. Considering values greater than +5 or smaller than -5 as a strong preference for
the specific learning style, learners were divided into two groups. The first group consists of
learners who have a strong learning style preference for at least one of the three
investigated learning style dimensions (N = 39). The second group consists of learners, who
have no strong preference for any of the three learning style dimensions (N = 33). After
testing whether data are normal distributed, t-test was applied, using a significance level of
0.05, in order to identify whether the learners from these two groups have significant
differences in their achievement.
9
Econf3-Bahrain University-April 6-8-2010
Fig.3 Pre-test operations for determining student's Learning Styles.
The evaluation of ILS depends on the balance of output (from 1 to 11) If the student's
answers rages from 1-3 then the student is fairly well balanced on the two dimensions
of that scale. If the total output of answers rages from 5-7, then the student has a
moderate preference for one dimension of the scale and will learn more easily in a
teaching environment which favors that dimension. And if the total output of answers
rages from 9-11, the student has a very strong preference for one dimension of the
scale. You may have real difficulty learning in an environment which does not
support that preference. On the other hand, our model's evaluation depends on the
evaluation of the answer. For example, if the student chose "A" as an answer for the
first question, this gives one point to the Activist pattern. Then we consider the
student's type of education in terms of his total number of points. Some questionnaire
questions that are available today to determine the student's learning style sometimes
lack clarity, which confuse the reader and the targeted questionnaire audience. For
that reason we tried to make the questionnaire as clear as possible, in addition to it
being suitable for the targeted audience, with great care for not being long or tedious,
because most of the questions in the questionnaires cause the reader to lose
10
Econf3-Bahrain University-April 6-8-2010
concentration during answering the questionnaire which affects the reader's answers.
For example, Dunn & Dunn questionnaire [18] consists of 118 questions, while our
questionnaire consists of 15 questions only. Fig.4 shows the resulted type for the user
model/profile.
Fig.4 the resulted type for the user model/profile.
6 The Proposed System Structure
Fig. 5 shows the architecture of the proposed system. It consists of several modules
that work together to produce an effective e-learning system. These models include:
System Administrator, Course Units Manager, Pre-test Statistics Analyzer, Exams
Generator, Chatting, News Manager, and Video Player. The System Administrator
checks students ID, name and password as a registered user or visitor. The Course
Units Manager manages adding and editing chapters and lessons and their order and
student's eligibility to browse the lesson or not. The Pre-test Statistics Analyzer
checks the questionnaire answers and gives learning styles results online. Exam
Generator module chooses a certain number of questions from unlimited multiple
choices questions and produces them for a student to answer. Exam Corrector corrects
the student answers and gives online marks. Chatting module allows student to
communicate together and with teachers. The News Manager module displays the last
four news about subject or website on the main screen of system. Finally, the Video
Player allows video education materials to be displayed in film format.
The system contains many modules. The main components of the model: pre-test for
determining learning style, the chatting, e-mails, online tests and Questions Generator.
The model uses the learning styles in order to reach the active adaptivity between all
contributing participants of e-learning system.
11
Econf3-Bahrain University-April 6-8-2010
The main contributing participants of an e-learning system are students (learners),
teachers (educators), system developers, administrators, and psychologists. In our
system, administrator have to be able to use HTML, PHP, MySQL, FrontPage and
FTP. We think that one of the important issues to improve the educational systems is
the supporting of personalization of learning goals, study content, learning pathways,
navigation, and assessment. As the previous explanation, In traditional or pedagogical
education, material to be learned is often transmitted to students by teachers. That is,
learning is passive. In active learning, students are much more actively engaged in
their own learning while educators take a more guiding role. This approach is thought
to promote processing of skills/knowledge to a much deeper level than passive
learning. Related terms/concepts include: experiential learning, hands on learning.
The electronic course provides the educational subject to the students according to the
learning style which will depend on a specific pre-test consists of 15 questions, Fig2,
before giving any lesson or unit. Understanding your learning styles is the first step in
improving your 'learning power' and gaining the most from the learning opportunities
your encounter.
The final exam consists of many random questions. The students will be given 20
questions on the web site on the basics of the educational subject, this is what we call
novice test. If the student passes this exam, he can go to the next exam which is for
intermediate, which is consists of 30 questions but more difficult than the first one.
After that, if the student passes the second test he can get in the final test which is for
advanced. This test consists of 40 more precise questions.
Fig5 components of the proposed Adaptive eLearning System
12
Econf3-Bahrain University-April 6-8-2010
In teaching, e-mail has served as a practical method to deliver actual information to
students or to give personal supervision. It has also been used to support national and
international communication between schools/universities located far away from each
other. Although the basic idea of e-mail is to serve as a tool for dyadic
communication, it can also be used in larger collaboration. With the help of mailing
lists, a larger group of students can use e-mail in sharing joint documents and in
commenting on each other’s work.
Chatting is a place for teachers to discuss and share ways that television and
technology can support education. Student can read what other people have already
said about several educational topics and add his own statement on the topic or he can
post a new query. By chatting, a great new online education forum, offering those
who have interest in education related topics an opportunity to share ideas and
insights. Chatting module is designed to facilitate communication between all who
share the common interest in quality education.
E-Learning uses electronic media to improve teaching and learning. Web-based
multimedia presentation is presenting the developed exposition at the most opportune
time. Once created, we can offer the web-based multimedia presentations to our
students repeatedly. For example, a given student may review a given module several
times. Likewise, students lacking the required knowledge are directed to review a
presentation on this knowledge area that might have been developed for the
prerequisite course. The course designer’s task is sequencing the reusable web-based
multimedia presentations for students.
7 Proposed Course Content Taxonomy
In order to enhance web-based courses, it is preferred to make the course material
richer and more flexible, so that different students can get personalized content, and a
personalized order of presentation. Some previous work was done in [32-34]. Fig.6
shows the proposed course content taxonomy used, where design of lesson content
is tailored to individual users, taking into consideration specific learning style,
analyzing coordination between student's learning style guidelines for preparing
learning materials according to different learner's characteristics.
Meeting the needs of the students is the cornerstone of an effective program of
"distance education", it is the index by which we judge where the effort that was made
reached in this area of progress. An additional advantage of the proposed system is
that, it takes up a massive number of teachers and learners efficiently and quickly,
since it contains a control panel for administrator to determine rules of the beneficiary
with more stability and predictability.
The system also allows the teacher to add chapters and lessons for the curriculum, as
it can also add tests for the learned lessons. Moreover it also allows the adding of an
unlimited number of questions for the tests to classify the question's level: beginner,
intermediate or advanced. It also allows the teacher to edit the personal data for
students to make data editing easier and faster. Teachers will use the system to enrich
or replace traditionally held lectures, supervise and support students, and simplify
everyday tasks like sending announcements, assigning exercises, or grading.
Comparison between the e-learning and the traditional learning, show that, the elearning will be effective when the methods and techniques used are suitable for the
13
Econf3-Bahrain University-April 6-8-2010
education process, when there is an interaction between the students with each other
and when the notes are exchanged between the students and the teacher from time to
time, and in the appropriate time. We made sure that the education system includes
all of the properties that would achieve these properties. This educational system, in
addition to the above, reveals the student's motivation style: Beginner, Intermediate
and Advanced.
Fig.6 AI Course Curriculum Content Taxonomy
8 System Implementation and Results
The tools which are used to implement the system and web site include: HTML, MS
Front Page, Photoshop, Animation, MySQL, and FTP. HTML (Hyper Text Markup
Language) is the glue that holds together every web site. MS Front Page to edit
HTML tags, PHP is the programming language to deal with server. The Photoshop is
used to design website's pages. The Multimedia is used to present some of lesson's
files. MySQL is the programming language that used to deal with database. FTP (File
Transfer Protocol) is the protocol that conveys the files on website.
The system main screen for the e-learning site contains icons and buttons and
the subject news. Those icons and buttons are: Lessons, references, LS determining
pre-test, exams with levels, e-mail to contact with administrator, forum, chatting, help
and suggestions. The learner login to enter the site as a registered student. The student
has to write his name and e-mail as a necessary condition to accept his registration.
14
Econf3-Bahrain University-April 6-8-2010
Fig. 7 shows the pre-test questionnaire screen for determining student's learning style.
The learner can't enter the lessons or exams before filling this page. This
questionnaire consists of fifteen questions, and every question has two options. Fig. 8
shows screen that displays the student's learning style based on his answers and shows
the weights of other styles based on pre-test statistics which are implemented by IF
statements.
Fig.7 the questionnaire Screen for determining student's learning style.
Fig.8 System Screen for results of LS questionnaire
15
Econf3-Bahrain University-April 6-8-2010
9 System Performance and Evaluation
9.1 Evaluation Process
Evaluation consists of numerous attributes; however it basically means assessing the
effectiveness and possible improvement of a course/curriculum. This can include
making sure that the learning process is liked by learners, and learners gained
information during this process. And assure that learners are accountable for the
information they obtained in the training. And assess learning outcomes, and find and
fix quality issues in the training as well as learn how to make training courses and
curriculums better in upcoming projects. Each of these reasons can improve the
training if fixed. For example, if after the evaluation the training is altered to make
sure the training is liked by the learners it can lead to an increase participation in the
training, an increase in learner retention, ensuring that it accommodates different
learning styles.
Fig. 9 shows the overall proposed system architecture which is built based on the Web
technology and its relationship with learners.
To increase the quality of training another critical reason for evaluation is to assess
the value of the training. This is essential because training is a part of business. If it is
not deemed valuable within the organization, then the amount spent on training is
often reduced. Reasons for evaluation in this particular area include adding value to
the organization, justifying the investment in training, assessing the effect that the
training has on profitability, the impact the training has on employee's work habits,
effectiveness and efficiency of the training, assessing the effect of customer
satisfaction from the training.
In evaluation learning we are searching for the things:
o Did the students like the training?
o Was the training relevant?
o Was the time spent by the students well spent?
16
Econf3-Bahrain University-April 6-8-2010
o Did the experience what was intended or something completely different?
o Was there a change in behavior due to the training and is the student aware of
the change?
The students' attitudes to the website are measured using a visual analogue
questionnaire. Without any information about the user, the adaptive system is notable
to adapt itself to the user’s characteristics and preferences. The statements in the scale
covered issues relating to the students' enjoyment of using the proposed environment.
The questionnaire consists of the following ten statements:
1. I enjoyed using the system.
2. I think the system made my learning worse.
3. I think I would like to use it again.
4. Using the website helped me to learn better.
5. The website makes learning easier.
6. I think there is no any technical problem with using the website.
7. I think I need someone to help me use the website.
8. A teacher is more helpful than the website.
9. The website is boring.
10. Using the website is not effective as the traditional learning.
With these questions we measured the performance of our model. However, we need
to determine the student's learning style first. Some researches [35,36] found that 48%
of students sample preferring to process visual information, 36% auditory and 16%
preferring kinesthetic and tactile instruction. Finally, the development of instructional
systems requires an interactive communication between the student and the tutor.
During running of our proposed system on the internet site, we got the numerical
results shown in Fig10. When teaching AI subject, results show that, the majority of
the students was Activist, while the minority of the students was Pragmatist.
Fig 10: Experimental results of the students' learning styles
9.2 Learning Gain
Some of the previous work results [37] show that the majority who accept e-learning
is 75% accept that e-learning is good as traditional learning, 10% did not accept,
while 15% remained undecided. According to Al-Zamil [38], 47.8% of his study
sample believed that the material costs to connect into the Internet form e-learning
barriers, and 44.4% of the respondents believe that the disadvantage of e-learning is
the lack of a professor when he/she is needed.
17
Econf3-Bahrain University-April 6-8-2010
The use of the t-test in [39] showed that there were no differences between members
of the sample which had both genders (males/females), and full-time students interact
more with the e-learning than part-time students, and the use of technology such as
computer and internet helps to interact with e-learning technology; as the response of
students who use computers is higher than the other students who do not use
computers. This study reported that 80.2 % of the sample did not agree on the
feasibility of e-learning, but in contrast, we find that 40.2% of them agree that the elearning method is suitable for education and we will prove its successfulness here.
The development of instructional systems requires an interactive communication
between the student and the tutor. For n = 27 students, we found with our system that:
51.85% of them are activists, 22.22% are reflectors, 22.22% are theorists and 3.71%
are pragmatists. Table5 summarizes these results.
Table 5 Experimental results of the students' learning styles
The success of an e-learning system can be considered as an emerging concept of
‘social issues’ and ‘technical issues’ and depends on numerous circumstances. Fig11,
Fig.12 show results of the students who learned using traditional method, and who
used the proposed eLearning system, respectively, when the students studied the same
AI subject. For a comparison between students who are taught using the proposed
eLearning system, we found practically that 30% of students achieved the excellent
degree, while 20% of students only achieved the excellent degree from those used the
traditional learning. And 3% only got low degree from students who taught using the
system, compared with 10% of students who used traditional learning.
These results prove that the proposed educational system is successful and has
contributed in improving the academic levels of students. As well as it illustrated the
importance of the adaptive e-learning system that based on learning styles. Fig.13
shows learners demand to our educational site during the period between January and
August 2009. Most access of the site was in April 2009.
18
Econf3-Bahrain University-April 6-8-2010
Fig.11
Fig12
Fig. 13: Results Traditional learning and E-learning
19
Econf3-Bahrain University-April 6-8-2010
9.3 System Scalability
Scalability as defined in [40] is a measure of an application system's ability to costeffectively provide increased throughput, reduced response time and/or support more
users, when hardware resources are added. For example, if the software architecture is
not able to use additional resources to increase throughput, the system will not be
scalable. We notice from Fig.14 that the system server gives a user more than 30
Mbps in perfect situation. This means that the system can accommodate 100 students
at the same time without falling. On August 12, at exactly 9:30 p.m., we asked thirty
students to enter the system together to test the availability of the system and we
found that the system was fine. Table 6 shows the number of unique visitors into our
web educational system site on the internet, and number of visits, and the band width
allocated during several months in 2009.
Fig. 14 Impact of changing the number of states.
Table 6 number of unique visitors to the system during year 2009.
20
Econf3-Bahrain University-April 6-8-2010
10 Conclusions
In this paper, a design and implementation of an Automated Adaptive Content Web
based E-Learning System (AACWELS) for teaching AI subjects is introduced. The
system uses an adaptive course content taxonomy based on a modified Felder's
questionnaire depending on the learning style properties used by Honey and
Mumford. The system tends to pursue adaptation according to obtained user profile.
Lessons content is tailored to individual users, taking into consideration a specific
learning style and subject matter learning motivation. System results show that the
number of excellent student who used the proposed system is increased to 30%, while
failure student is reduced to 3%, while students who used the traditional education
method achieves 20% excellent students, and 10% failure students, when students
studied the same AI subject . These results prove that student learning gain was raised
in comparison with other systems and with traditional learning methods. We also
proved practically that the system is scalable and can accommodate about 100
students at the same time without falling.
REFERENCES
[1] Abdul Hamid M. Ragab & Abubakr Salim Bajnaid, An Effective-Adaptive Elearning System Based on Multi-Styles Assessmen, King AbdulAziz
University,Jeddah, Saudi Arabia, the 7nth Annual Symposium on Learnining and
Technology, Edutainment Effat Univ. 10-11 June 2009
[2] Abdul Hamid M. Ragab, (2005). "A New Adaptive e-learning Multimedia Styles
'Web Based Model' for Teaching Computer Science Subjects", Proceedings of
Workshop Means to Implement the Document of Views of Prince Abdullah Bin
Abdulaziz on Higher Education, Dept. Of Computer Science, Faculty of Science,
King AbdulAziz University, Jeddah, Saudi Arabia, 2005.
[3] Abdul Hamid M. Ragab " Multi-Styles Adaptive e-Learning for Teaching &
Learning Holy Quran with Quality Services", Accepted for publications in ‫ندوة لقردن‬
– ‫ لقموينة لقمنور‬2009 ‫ أكتوبن‬- ‫لقكنيم ةلقترنيات لقمعاصن‬
[4] Herman Dwi Surjono,"The Development of an adaptive eLearning System based
on learning style Diversity of Visual-Auditory-Kinesthetic", International Seminar on
ICT for Education , Yogyakarta State University, Indonesia, 13–14 February, 2009.
[5] E. Triantafillou, A. Promortsis, and E. Georgiadou. (2002). AES-CS: Adaptive
Educational System base on cognitive styles. In Proceedings of the AH2002
Workshop, Malagas, Spain, pp. 10-20.
[6] LSAS.,http://www.archives.ecs.soton.ac.uk/users/nb99r/intro_short/
frame.htm.
[7] Cs388, Carver, C.A., Howard, R.A., and Lavelle, E., Enhancing student learning
by incorporating learning styles into adaptive hypermedia.,In Proceedings of EDMEDIA ’96,World Conf. on Educational Multimedia and Hypermedia,(Boston, USA,
1996), 118-123.
21
Econf3-Bahrain University-April 6-8-2010
[8]Gilbert, J.E. and Han, C.Y.
Adapting instruction in search of ‘a significant difference’. Journal of Network and
Computer applications, 22, (1999).
[9] C. Wolf. (2004). iWeaver: Towards 'Learning Style'-based e-Learning in
Computer
Science
Education,
available
online
at:
http://www3.cti.ac.at/icl/archive/presentation/wolf.pdf.
[10] Stern, M. and Woolf, P. Adaptive content in an online lecture system,
Proceedings of the International Conference on Adaptive Hypermedia and Adaptive
Webbased systems (Trento, Italy, 2000), 291-300.
[11] Felder, R.M. and Silverman, L.K. 1988. Learning and Teaching Styles in
Engineering Education. Engineering Education. 78, 7 (1988), 674--681.
http://chat.carleton.ca/~tblouin/Felder/felder.html.
http://virtualschooling.wordpress.com/2009/02/12/using-learning-styles
[12] Rosa María Carro, Estrella Pulido, Pilar Rodríguez, TANGOW: a Tool for
Designing Adaptive Web-based Courses.,Proceedings of the 2nd Workshop on
Adaptive Systems and User Modeling on the WWW,
wwwis.win.tue.nl/asum99/carro/carro.html
[13] M. Grigoriadou, K. Papanikolaou, H. Kornilakis, and G. Magoulas. (2001).
INSPIRE: An Intelligent System for Personalized Instruction in a Remote
Enironment. In Proceedings of the 3rd Workshop Conference on Adaptive Hypertext
and Hypermedia, Sonthofen, Germany, pp. 13-24.
[14] P. Brusilovsky and C. Peylo. (2001). Adaptive hypermedia. User Modeling and
User-Adapted Interaction, 11, pp. 87-110.,
also AHA!, http://aha.win.tue.nl.
[15]John Eklund and Peter Brusilovsky, InterBook: an Adaptive Tutoring System,
Carnegie
Mellon
University,
http://science.uniserve.edu.au/newsletter/vol12/eklund.html.
[16] Karan Sharma, Designing Knowledge Based Systems as Complex Adaptive
Systems , http://ai.uga.edu/karan123/AGI_paper.pdf.
[17] D Del Corso, 3DE: AN ENVIRONMENT FOR THE DEVELOPMENT OF.
LEARNER-CENTERED
CUSTOM
EDUCATIONAL
PACKAGES,2001,
ieeexplore.ieee.org/iel5/7628/20803/00963703.pdf
[18] R. Dunn, and K. Dunn, (1978). Teaching Students through their individual
learning styles: A practical approach. Restone, VA: Restone Publishing.
[19] P. Honey, and A. Mumford. (2006). Learning Styles Questionnaire, 80-item
version. Maidenhead, UK, Peter Honey Publication Ltd.
[20 ]Chris Jackson ,"A programme for all learning styles": the Honey & Mumford
learning styles,Feb 2009 www.trainingzone.co.uk/cgi-bin/item.cgi.
[21] R. Felder, and J. Spurlin. (2005). Applications, Reliability and Validity of the
Index of Learning Styles. Int. J. Engng. Education, Vol. 21, No. 1, pp. 103-112.
22
Econf3-Bahrain University-April 6-8-2010
[22] R. Felder. (1988). Learning and Teaching Styles in Engineering Education, Engr.
Education, 78(7), pp. 674–681. http://www.ncsu.edu/felder-public/Papers/LS1988.pd.
[23] G. Weber, and P. Brusilovsky. (2001). ELM-ART: An Adaptive Versatile
System for Web-based Instruction. International Journal of Artificial Intelligence in
Education,12,
4,
pp.351-384,
available
online
at
http://cbl.leeds.ac.uk/ijaied/abstracts/Vol_12/weber.html.
[24] T Kubeš ,Overview of Existing Adaptive Hypermedia e-Learning Systems,
www.tomaskubes.net/.../kubes-overview_of_existing_adaptive_hypermedia_elearning_systems.pdf
[25] Ricardo Conejo,etal, SIETTE: A Web–Based Tool for Adaptive Testing,
International Journal of Artificial Intelligence in Education 14 (2004) 1 -33
IOS Press, 2004 – IOS Press.
[26] J.M. López, E. Millán, J.L. Pérez-de-la-Cruz & F. Triguero, ILESA: aWeb-based
Intelligent Learning Environment for the Simplex Algorithm,
http:// citeseer.ist.psu.edu/521998.html ttp://
[27] D. Kolb. (1984). Experiential Learning: Experience as the Source of Learning
and Development. Prentice-Hall, Inc., Englewood Cliffs, N.J.
[28 ]Doreen J. Gooden ,"An Examination Of Kolb’s Learning Style Inventory", 2009
ABR
&
TLC
Conference
Proceedings.
Oahu,
Hawaii,
USA.
www.cluteinstitute.com/Programs/Hawaii_2009/
[29 ]Rationale and Strategies for Using Kolb Learning Style Theory:, March, 2009 ,
https://www.novapublishers.com/catalog/product_info.php.
[30] S Visser ,Teaching Styles versus Learninig Styles: a Comparative
Analysis,www.wmin.ac.puk/pdf/ - 2006
[31] C. Froschl. (2006). 'User Modeling and User Profiling in Adaptive E-learning
Systems: An approach for a service-based personalization solution for the research
project AdeLE (Adaptive e-Learning with Eye-Tracking)', Institute for Information
Systems and Computer Media (IICM), Faculty of Computer Science, Graz University
of Technology, Graz University of Technology.
[32] M. Stern, and P. Woolf. (2000). Adaptive content in an online lecture system. In
Proceedings of the international Conference on Adaptive Hypermedia and Adaptive
Web-based systems, Trento, Italy, pp. 291-300
[33] P. Paredes, and P. Rodrigues. (2002). Considering sensing-intuitive dimension to
exposition-exemplification in Adaptive Sequensing. In Proceedings of the AH2002
Conference, Malaga, Spain, pp.556-559
[34] D. Milosevic, M. Brkovic, and D. Bjekic. Designing lesson content in adaptive
learning environment. In Proceedings of Conference IMCL2006, Amman, Jordan, 1921 April, 2006.
[35] H. Surjono. (2009)."The Development of an adaptive eLearning System based on
learning style Diversity of Visual-Auditory-Kinesthetic", International Seminar on
ICT for Education , Yogyakarta State University, Indonesia, 13–14 February, 2009.
23
Econf3-Bahrain University-April 6-8-2010
[36] Alexandra I. CRISTEA, A Teaching & Learning Framework for E-Learning
Systems: Application of Learning & Perceptual Styles, wwwis.win.tue.nl/ ~acristea/
WBE/416-024omg.pdf
[37] Vitalis Ndume, CHALLENGES OF ADAPTIVE ELEARNING AT HIGHER
LEARNING INSTITUTIONS: A CASE STUDY IN TANZANIA, University of Dar
Es Salaam, http://ijcir.org/volume2-number1/article6.pdf.
[38]Z-Alzamil,
‫ترييم تجنبة لقتعليم لإلقكتنةني في بعض مؤسسات لقتعليم لقعاقي باقمملكة لقعنبية لقسعودية من ةجهة نظن‬
،‫ لقكلية لقترنية باقنياض‬،‫ قسم ترنية لقحاسب‬،‫لقطاقب‬
zakarya@computer.org
[39] Student's t-test, of http://en.wikipedia.org/wiki/Student's_t-test, 2009.
[40] L. G. Williams, and C. U. Smith. (2005). QSEM: Quantitative Scalability
Evaluation Method, Proceeding of the Computer Measurement Group, Las Vegas.
24