AN ADAPTIVE AND INTELLIGENT TUTOR BY EXPERT SYSTEMS FOR MOBILE DEVICES

International Journal of Managing Public Sector Information and Communication Technologies (IJMPICT)
Vol. 3, No. 1, September 2012
DOI : 10.5121/ijmpict.2012.3102 21
AN ADAPTIVE AND INTELLIGENT TUTOR BY
EXPERT SYSTEMS FOR MOBILE DEVICES
Hossein Movafegh Ghadirli1
and Maryam Rastgarpour2
1
Graduate student in Computer Engineering, Young Researchers Club, Islamshahr
Branch, Islamic Azad University, Islamshahr, Iran
hossein.movafegh@iau-saveh.ac.ir
2
Faculty of Computer Engineering, Department of Computer, Science and Research
Branch, Islamic Azad University, Saveh, Iran
m.rastgarpour@iau-saveh.ac.ir,m.rastgarpour@gmail.com
ABSTRACT
Mobile Learning (M-Learning) is an emerging discipline in the area of education and educational
technology. So researchers are trying to optimize and expanding its application in the field of education.
The aim of this paper is to investigate the role of mobile devices and expert systems in disseminating and
supporting the knowledge gained by intelligent tutors and to propose a system based on integration of
intelligent M-Learning with expert systems. It acts as an intelligent tutor which can perform three
processes - pre-test, learning concept and post-test - according to characteristic of the learner. The
proposed system can improves the education efficiency highly as well as decreases costs. As a result,
every time and everywhere (ETEW) simple and cheap learning would be provided via SMS, MMS and so
on in this system. The global intention of M-Learning is to make learning “a way of being”.
KEYWORDS
Expert Tutor, Intelligent Tutoring System, Mobile Learning, Adaptive Learning
1. INTRODUCTION
Nowadays, researchers try to develop a new paradigm of e-learning to mobile space(M-
Learning). The term “Mobile Learning” means educational application of a variety of mobile
technologies. M-Learning is so regarded after advent of computer-based learning. But its
implementation is more complex than computer-based learning [1]. Mobile devices and wireless
networks can provide texts, video on demand, and information services. Learners would require
such technology as appropriate mobile devices, and high-bandwidth wireless networks.
Connectivity is also important because disconnection caused by handovers seriously affect the
quality of information services.
Diversity of learners leads to decrease the efficiency of this style. In fact, some lessons must be
repeated several times for a number of learners while the same courses may be removed for
others [2].
Later, the researchers in pedagogy sciences concluded that the learning must be dynamic and
intelligent. The fact is that an expert tutor can adapts lessons sequence and training speed with
aptitude and characteristics of learner. He can also adjust the expression style with learner’s
mood as well as cancels the class due to incorporate mental conditions of the learner.
The most of recent systems run on desktop computers and are not designed for mobile
application such as mobile phones, smart phones, PDAs, etc. On the other hand, M-Learning

22
systems aim to improve the quality of learning by providing mobile learners with an easy,
contextualized and ubiquitous access to knowledge [1].
M-Learning has some problems, like technical nature, which is related to porting existing
systems to a new environment. It introduces some complex implementation issues. Namely, the
wireless environment (GSM, UMTS, WLAN) is much different from the standard "wireline"
one; bandwidth, delay, error rate, interference and the like, may change dramatically as the
learner changes her/his location [1].
This paper presents an intelligent system to apply the abilities of expert systems. So E-learning
would be efficient, adaptive and performed by mobile devices. Adaptation is very important,
because the contexts would be used by millions variant learners. So the concept, which is
developed for one user, isn’t applicable for others [3].
The proposed system determines the learning style via a test. Then the learning process starts.
Gradually, some characteristics of learner may be change by learner's progress. These
improvements would be saved by system in learning process. So learner model gets closer step
by step. The system can receives scientific and mental feedback of learner expertise and then
change the learning style during the process. M-Learning has surpassed of Web-based learning
because, it is estimated the total number of mobile phone users worldwide is over 300 million,
double the number of Internet users [4]. M-Learning content is installed and supported in one
place while millions learners can use it just via a mobile device. The aim of proposed system is
to offer the content which the user is not aware of it.
The rest of this paper is organized as follows. Section 2 defines an intelligent tutoring system
and presents some available samples. Then it deliberates M-Learning and some learning styles
in section 3. Section 4 describes the proposed M-Learning system which is intelligent, adaptive
and customizable. Finally this paper concludes in section 5.
2. INTELLIGENT TUTORING SYSTEM (ITS)
ITSs are computer-based instructional systems with educational content models. They specify
what to teach, and also teaching strategies that specify how to teach [5]. In late 1960, the ITSs
have moved out of academic labs and have been applied in classrooms and workplaces. Some of
them have shown to be high effective [6]. Unfortunately, intelligent tutors are difficult and
expensive to build whereas they are more common and have proven that to be more effective.
Intelligent systems can recognize the learner type, choose appropriate course content from
knowledge base and present it to learners in proper style [2]. It also attempts to simulate a
human tutor expertly and intelligently. Students using these systems usually solve problems and
related sub-problems within a goal space, and receive feedback when their behaviour diverges
from that learner model.
Some design factors of ITS include of component expert simulator, tutor software, learner
model, modeller, and knowledge base. Fig. 1 illustrates them.
ITSs allow “mixed-initiative” tutorial interactions, where learners can ask questions and have
more control over their learning.
A mobile ITS integrates three fields of research that contribute to the design and
implementation of the mobile intelligent tutoring system, namely the Intelligent Tutoring
Systems, Mobile Human Computer Interaction (HCI), and Mobile Learning disciplines, as
shown in Fig. 2.[7]

23
Figure 1. Intelligent Tutoring System Components [2]
Some available intelligent tutoring systems are introduced in the following.
xTEx-Sys [8] is actually not an intelligent tutoring system itself; it is an authoring shell
supporting the development of ITSs.
Älykkö [9] works as a basis for M-Learning and a tutoring dialogue between students and
teachers. In addition, it enables the creation of the structure and framework for M-Learning
process taking place in authentic environments. Älykkö is a PHP- and MySQL-based
application with automatic and semiautomatic tutoring features.
C-POLMILE [10] is an intelligent learning environment with an open learner model for C
programming. There are two versions of the system - the desktop PC version, and a version for
use on a handheld computer. The student can use whichever is most convenient at the time, and
can take part in the kind of interaction which is most appropriate for their current location, the
device being used, the time they have available, the likelihood of distraction in that location, etc.
Figure 2. Integration of disciplines for the Mobile Intelligent Tutoring System [7]
The next section describes the properties of mobile devices and It also express how to learn by
such devices.

24
3. MOBILE LEARNING
Mobile learning is an emerging discipline in the area of education. A review of M-Learning
projects funded by the European Union since 2001 [11]. It confirms that mobile phones are the
most frequently used device in these projects, followed by PDAs and other handhelds, with
personal listening devices (e.g. iPods) receiving a little less attention. M-Learning is the point at
which mobile computing and electronic learning intersect to produce an anytime, anywhere
learning experience.
The field of M-Learning is approximately a decade old and has grown rapidly. As personal
computer integration predicated computer-enhanced learning, a primary factor in the growth of
M-Learning is the increasingly ubiquitous integration of cell phones into society [7].
The term “Mobile Devices” includes standard cell phones (those are without an operating
system performing basic cellular voice communications), smart phones (those utilize an
operating system providing voice services as well as additional data processing applications),
and personal digital assistants (PDAs provide data processing without voice capabilities). User
interaction in laptop computers is similar to desktop computers. Laptop computers are portable
while desktop computers are not. These reasons lead to increase laptop users.
This system for M-Learning defines several functional levels, simplifying the design and
development. So that different parties (e.g. vendors, providers, designers) can address individual
levels [12]. This system enjoys all entities. So a single entity is not forced to do all tasks for
developing an M-Learning system in this way.
The system is divided into four levels. In the M-Learning applications level, many new
applications gets possible, and many existing electronic learning applications can be modified
for a mobile environment. In the mobile user infrastructure level, the design of new M-Learning
applications should consider the capabilities of the user mobile devices. In the mobile protocol
level, the aim is to hide the underlying network's details from applications while providing a
uniform and easy-to-use interface. In the mobile network infrastructure level, service quality
primarily depends on network resources and capabilities.
3.1. Learning Styles
It’s worth to be noted that the concept of “learning style” has emerged as a focal point of much
psychological research. In our proposed learner model, we have adopted the Jackson’s learning
styles profiler in order to model learning styles of the learners. Jackson proposed five learning
styles which are summarized in Table 1. Reader can find more details in [2].
Table 1. Summarization of the available Learning Styles [13]
Learning style Comment
Sensation Seeking (SS) Believing that the experiences create learning.
Goal Oriented Achievers (GOA) Self-confident to achieve difficult and certain target.
Emotionally Intelligent Achievers (EIA) Rational and goal-oriented.
Conscientious Achievers (CA) Responsible and insight creator.
Deep Learning Achievers (DLA) Interested in learning highly.
4. PROPOSED SYSTEM
The adaptive and intelligent tutor by expert systems is to promote the reading ability of learners
based on the individual learning styles. It can estimate the reading abilities of individual
learners. Then it recommends appropriate content to individual learners according to learner’s

25
simple feedback responses and difficulty parameters of the contents. It can determine learner
type (especially in terms of learning style), learning content adaptively. So that it can be updated
automatically with learner’s characteristics and behaviour in server side. Learning all courses
can be customized well at home via SMS, MMS and so on in this system. So learners can solve
some examples and proper exercises ETEW. Finally he can attend in course examination which
would be virtual or physical.
4.1. The Interface
Intelligent tutors, such as Andes, Algebra I, and AutoTutor, provide users interfaces divided into
regions. Andes, an intelligent tutor for Newtonian Physics, have an interface divided into three
regions/panes [14]. Similarly, Algebra I and AutoTutor developers employed a multi-region
interface design [15, 16]. Users are given the specific problem or scenario which they are
currently working on, a workspace, hint buttons, and other areas such as a glossary, or reference
material necessary. As students use the tutors on large desktop displays, they are easily able to
see all of these regions without having to manually navigate to each individually. The problem
of tutor implementation in the smaller device is that the displays are small as well.
The mobile users would like to access some information without keyboard and typing the words
by small fonts in a mobile tutor interface.
When the user is interacting with the mobile tutor, he should solve the problems instead of
getting help. So user time is wasted. It seems that there is some interface challenges such as the
limitation in user input in addition of the problems in small displays [17]. However, we focus
the mobile interface.
4.2. Training Method
To find the role of a mobile tutor in learning, we will examine how mobile tutors can be
integrated into classes and the education. The primary difference between mobile tutor and
desktop tutor is the frequency and duration of applying. Now, two questions should be
considered. The first question is “Is it practical to use a small size screen and to limit interaction
modes?” . Another one is whether the M-Learning is as efficient as the desktop counterparts in
terms of frequency and duration. However, considering that the students are often from the 5th
and 6th grades, they comfortable play video games through small handheld consoles and we
expect that they would be equally comfortable when using a mobile tutor [18].
Expert tutor’s knowledge includes of two parts, course knowledge (learning content) and
learning technique. Course knowledge is theoretic information, technical content and probably
experiments which expert tutor notes. Learning technique is some experiences which he have
got during teaching years [19]. An expert tutor determines learner level in according to IQ,
understanding, behaviours, talent and individual characteristic like physical class. Learner level
consists of “weak”, “slow learner”, “smart”, “genius” and so on. Tutor teaches educational
content corresponding to learner level in proper method such as film, dynamic view, and game.
So learner level may be changed while he get feedback from learner during training. The tutor
helps learner to learn by “the best way” in proposed system.
The expert tutor offers an education method based on learner’s type. Tutor often determine
different scores for variant sections according to education method. Moreover, he marks highest
score to the most important section of course in all education methods.
The smallest part of any topic which can’t divide more is called “concept”. It is usually
equivalent a lesson in physical class. Educational concepts transfer to knowledge base in this
system. Then the system can distinguish all concepts and relocate all parts. Sometimes, a lesson
is needed to repeat, relocate or even remove for a learner. Most of available systems guide a

26
learner to a special aim intelligently in learning process. While only a few intelligent systems
provides selecting subsections of a concept for a learner.
This system uses a three layered structure to offer and implement a concept:
1. Pre-test
2. Learning concept
3. Post-test
The pre-test includes of some questions planned by an expert tutor to determine learner’s
primary knowledge level. The learning concept depends on learner level. So the best method to
train a learner is determined in this way. Then learning process starts up. After learning is done,
a post-test evaluate the learner by some questions. Figure 3 shows architecture of proposed
system.
Figure 3. The Architecture of Proposed System
4.3. Learning Evaluation
Learner evaluation is significant. It has two levels, conceptual and objective. Evaluating in
concept level refers to learner understanding of lesson concept and evaluating in objective level
denotes to learner understanding of lesson topic [2]. Knowledge level of learner is determined
with concept level and objective level. The tutor can extract proper questions from question
base through an expert system, pre-test and post-test. He notes that a specific score is given to
each question.
Selecting question should satisfy some rules. Firstly, the questions should not be repetitious
even if a learner would be trained one concept several times. Secondly, the question must be
planned for all sections of a concept entirely and uniformly. Thirdly, expert tutor plans
questions in all level. Sequence, number and level of questions are determined according to
learner level and learning type intelligently. Sum of scores is calculated and learner level is
determined after answering the questions.
Table 2 presents five categories of learner’s knowledge level about a concept [20]:

27
Table 2. Categories of Knowledge Level
Knowledge Level Score
Excellent 86-100
Very good 71-85
Good 51-70
Average 31-50
This system updates the learner’s model during progress of question answering. This system can
also save last academic status of learner and all his learning records.
5. CONCLUSIONS
An adaptive and intelligent tutor by expert system for mobile devices was presented in this
research. The focus of this paper is to outline the process involved in studying the role of mobile
devices in disseminating and supporting the knowledge gained by intelligent tutors. The
proposed system can adapt with learning styles (i.e. Sensation Seeking, Goal Oriented
achievers, Emotionally Intelligent Achievers and Conscientious Achievers), aptitude,
characteristics and behaviours of a learner. It acts as an intelligent tutor which can perform three
processes - pre-test, learning concept and post-test - according to characteristic learner. This
system uses expert simulator and its knowledge base at server side. It is also based on SMS,
MMS and so on. It leads to be simple learning, low-cost, available everywhere and every time.
Consequently thousands of students can learn simultaneous and integrated efficiently.
Despite of facilitating all schools with Internet and computer access to students, a deeper
examination reveals that the presence of technology does not guarantee that it is applied
efficiently. Furthermore, it solves the drawback of previous system and human expert tutor. It
can improve efficiency of pedagogy and education too. In other words, it helps learners to study
in “the best way”.
REFERENCES
1. V. Glavinic, et al., “Agents in m-learning systems based on intelligent tutoring,” Universal
Access in Human-Computer Interaction. Applications and Services, pp. 578-587, 2007.
2. H. Movafegh Ghadirli and M. Rastgarpour, “A Web-based Adaptive and Intelligent Tutor by
Expert Systems,” in The Second International Conference on Advances in Computing and
Information Technology (ACITY 2012), Chennai, India, 2012.
3. Brusilovsky, P.: Methods and techniques of adaptive hypermedia. User Modeling and User-
Adapted Interaction 6, 2-3 (1996) 87-129.
4. F. Muller-Veerse, “Mobile Commerce Report”, Technical Report, Durlacher Research Ltd,
London, Nov 1999.
5. S. Ohlsson, “Some Principles of Intelligent Tutoring,” In Lawler & Yazdani (Eds.), Artificial
Intelligence and Education, vol. 1, Ablex: Norwood, NJ, pp. 203-238, 1987.
6. P. Brusilovsky, “Methods and techniques of adaptive hypermedia,” User Modeling and User-
Adapted Interaction, vol. 6, pp. 87-129, 1996.
7. Q. Brown, "Mobile intelligent tutoring system: moving intelligent tutoring systems off the
desktop," Drexel University, 2009.

28
8. V. Glavinić, et al., "Extending intelligent tutoring systems to mobile devices," 2008, pp. 245-
252.
9. A. Rytkönen, “An Intelligent Mobile Tutoring Tool Enabling Individualisation of Students’
Learning Processes.”
10. S. Bull, et al., “Learner models to promote reflection in combined desktop PC/Mobile intelligent
learning environments,” 2003.
11. A. Pęcherzewska and S. Knot, “Review of existing EU projects dedicated to dyslexia, gaming in
education and m-learning,” WR08 Report to CallDysc project, June 2007.
12. C. H. Leung and Y. Y. Chan, “Mobile learning: A new paradigm in electronic learning,” 2003,
pp. 76-80.
13. H. Movafegh Ghadirli and M. Rastgarpour, “A Model for an Intelligent and Adaptive Tutor
based on Web by Jackson’s Learning Styles Profiler and Expert Systems”, Lecture Notes in
Engineering and Computer Science: Proceedings of The International MultiConference of
Engineers and Computer Scientists 2012(IMECS 2012), 2012. 1: p.63-67.
14. A. S. Gertner and K. VanLehn, “Andes: A Coached Problem Solving Environment For Physics,”
in 5th International Conference on Intelligent Tutoring Systems, 2000, pp. 133-142.
15. A. C. Graesser, G. T. Jackson, and B. McDaniel, “AutoTutor holds conversations with learners
that are responsive to their cognitive and emotional states,” Educational Technology, vol. 47, pp.
19-22, 2007.
16. K. R. Koedinger and V. Aleven, “Exploring the Assistance Dilemma in Experiments with
Cognitive Tutors,” Educational Psychological Review, vol. 19, pp. 239-264, 2007.
17. Q. Brown, et al., “The design of a mobile intelligent tutoring system,” 2008.
18. L. Razzaq, M. Feng, G. Nuzzo-Jones, N. T. Heffernan, K. R. Koedinger, B. Junker, S. Ritter, A.
Knight, E. Mercado, T. E. Turner, R. Upalekar, J. A. Walonoski, M. A. Macasek, C. Aniszczyk,
S. Choksey, T. L. Ak, and K. Rasmussen, “The Assistment Project: Blending Assessment and
Assisting,” in 12th Artificial Intelligence in Education, Amsterdam, 2005, pp. 555-562.
19. Excellent Classroom Management, by: Rinne, C.H., Wadsworth Publishing Company, 1997.
20. El-Khouly M. M., Far B. H. & Koono Z., “Expert tutoring system for teaching computer
programming languages”, Expert Systems with Applications, 18, 27-32, 2000.
Authors
Hossein Movafegh Ghadirli received his B.S. in Computer Engineering from
Saveh branch, Islamic Azad University (IAU), Saveh, Iran in 2009 and He is
currently a graduate student in Computer Engineering at Science and Research
branch, IAU, Saveh, Iran. His overriding interest has been bringing E-Learning,
M-Learning and Intelligent Tutoring Systems to improve their productivity for
both government and commercial organizations. He is a member of Young
Researchers Club, Islamshahr Branch, Islamic Azad University, Islamshahr, Iran.
Maryam Rastgarpour received her B.S. in Computer Engineering from
Kharazmi University, Tehran, Iran in 2003, and the M.S. in Computer
Engineering from Science and Research branch, Islamic Azad University (IAU)
, Tehran, Iran in 2007.She is currently a Ph.D. candidate in AI there. She is also a
lecturer at Computer Department, Faculty of Engineering, Saveh branch, IAU for
graduate and undergraduate students. Her research interests include in the areas
of Machine Learning, Pattern Recognition, Expert Systems, E-Learning,
Machine Vision, specifically in image segmentation and Intelligent Tutor
System.

AN ADAPTIVE AND INTELLIGENT TUTOR BY EXPERT SYSTEMS FOR MOBILE DEVICES

More Related Content

What's hot (18)

Similar to AN ADAPTIVE AND INTELLIGENT TUTOR BY EXPERT SYSTEMS FOR MOBILE DEVICES (20)

Recently uploaded (20)

AN ADAPTIVE AND INTELLIGENT TUTOR BY EXPERT SYSTEMS FOR MOBILE DEVICES