Abstract: This paper presents an exact pruning algorithm with
adaptive pruning interval for general dynamic neural networks
(GDNN). GDNNs are artificial neural networks with internal dynamics.
All layers have feedback connections with time delays to the
same and to all other layers. The structure of the plant is unknown, so
the identification process is started with a larger network architecture
than necessary. During parameter optimization with the Levenberg-
Marquardt (LM) algorithm irrelevant weights of the dynamic neural
network are deleted in order to find a model for the plant as
simple as possible. The weights to be pruned are found by direct
evaluation of the training data within a sliding time window. The
influence of pruning on the identification system depends on the
network architecture at pruning time and the selected weight to be
deleted. As the architecture of the model is changed drastically during
the identification and pruning process, it is suggested to adapt the
pruning interval online. Two system identification examples show
the architecture selection ability of the proposed pruning approach.
Abstract: Markov games can be effectively used to design
controllers for nonlinear systems. The paper presents two novel
controller design algorithms by incorporating ideas from gametheory
literature that address safety and consistency issues of the
'learned' control strategy. A more widely used approach for
controller design is the H∞ optimal control, which suffers from high
computational demand and at times, may be infeasible. We generate
an optimal control policy for the agent (controller) via a simple
Linear Program enabling the controller to learn about the unknown
environment. The controller is facing an unknown environment and
in our formulation this environment corresponds to the behavior rules
of the noise modeled as the opponent. Proposed approaches aim to
achieve 'safe-consistent' and 'safe-universally consistent' controller
behavior by hybridizing 'min-max', 'fictitious play' and 'cautious
fictitious play' approaches drawn from game theory. We empirically
evaluate the approaches on a simulated Inverted Pendulum swing-up
task and compare its performance against standard Q learning.
Abstract: Mobile Learning (M-Learning) is a new technology
which is to enhance current learning practices and activities for all
people especially students and academic practitioners UTP is
currently, implemented two types of learning styles which are
conventional and electronic learning. In order to improve current
learning approaches, it is necessary for UTP to implement m-learning
in UTP. This paper presents a study on the students- perceptions on
mobile utilization in the learning practices in UTP. Besides, this
paper also presents a survey that was conducted among 82 students
from System Analysis and Design (SAD) course in UTP. The survey
includes basic information of mobile devices that have been used by
the students, opinions on current learning practices and also the
opinions regarding the m-learning implementation in the current
learning practices especially in SAD course. Based on the results of
the survey, majority of the students are using the mobile devices that
can support m-learning environment. Other than that, students also
agreed that current learning practices are ineffective and they believe
that m-learning utilization can improve the effectiveness of current
learning practices.
Abstract: Music segmentation is a key issue in music information
retrieval (MIR) as it provides an insight into the
internal structure of a composition. Structural information about
a composition can improve several tasks related to MIR such
as searching and browsing large music collections, visualizing
musical structure, lyric alignment, and music summarization.
The authors of this paper present the MTSSM framework, a twolayer
framework for the multi-track segmentation of symbolic
music. The strength of this framework lies in the combination of
existing methods for local track segmentation and the application
of global structure information spanning via multiple tracks.
The first layer of the MTSSM uses various string matching
techniques to detect the best candidate segmentations for each
track of a multi-track composition independently. The second
layer combines all single track results and determines the best
segmentation for each track in respect to the global structure of
the composition.
Abstract: Imitation learning is considered to be an effective way of teaching humanoid robots and action recognition is the key step to imitation learning. In this paper an online algorithm to recognize
parametric actions with object context is presented. Objects are key instruments in understanding an action when there is uncertainty.
Ambiguities arising in similar actions can be resolved with objectn context. We classify actions according to the changes they make to
the object space. Actions that produce the same state change in the object movement space are classified to belong to the same class. This allow us to define several classes of actions where members of
each class are connected with a semantic interpretation.
Abstract: In recent years, real estate prediction or valuation has
been a topic of discussion in many developed countries. Improper
hype created by investors leads to fluctuating prices of real estate,
affecting many consumers to purchase their own homes. Therefore,
scholars from various countries have conducted research in real estate
valuation and prediction. With the back-propagation neural network
that has been popular in recent years and the orthogonal array in the
Taguchi method, this study aimed to find the optimal parameter
combination at different levels of orthogonal array after the system
presented different parameter combinations, so that the artificial
neural network obtained the most accurate results. The experimental
results also demonstrated that the method presented in the study had a
better result than traditional machine learning. Finally, it also showed
that the model proposed in this study had the optimal predictive effect,
and could significantly reduce the cost of time in simulation operation.
The best predictive results could be found with a fewer number of
experiments more efficiently. Thus users could predict a real estate
transaction price that is not far from the current actual prices.
Abstract: The purpose of this research aims to discover the
knowledge for analysis student motivation behavior on e-Learning
based on Data Mining Techniques, in case of the Information
Technology for Communication and Learning Course at Suan
Sunandha Rajabhat University. The data mining techniques was
applied in this research including association rules, classification
techniques. The results showed that using data mining technique can
indicate the important variables that influence the student motivation
behavior on e-Learning.
Abstract: In the present study, a support vector machine (SVM) learning approach to character recognition is proposed. Simple
feature detectors, similar to those found in the human visual system, were used in the SVM classifier. Alphabetic characters were rotated
to 8 different angles and using the proposed cognitive model, all characters were recognized with 100% accuracy and specificity.
These same results were found in psychiatric studies of human character recognition.
Abstract: Emotions are related with learning processes and
physiological signals can be used to detect them for the
personalization of learning resources and to control the pace of
instruction. A model of relevant emotions has been developed, where
specific combinations of emotions and cognition processes are
connected and integrated with the concept of 'flow', in order to
improve learning. The cardiac pulse is a reliable signal that carries
useful information about the subject-s emotional condition; it is
detected using a classroom chair adapted with non invasive EMFi
sensor and an acquisition system that generates a ballistocardiogram
(BCG), the signal is processed by an algorithm to obtain
characteristics that match a specific emotional condition. The
complete chair system is presented in this work, along with a
framework for the personalization of learning resources.
Abstract: Malaysia government had been trying hard in order to
find the most efficient methods in learning. However, it is hard to
actually access and evaluate students whom will then be called an
excellent student. It is because in our realties student who excellent
is only excel in academic. This evaluation becomes a problem
because it not balances in our real life interm of to get an excellent
student in whole area in their involvement of curiculum and cocuriculum.
To overcome this scenario, we designed a module for
Student Idol to evaluate student through three categories which are
academic, co-curiculum and leadership. All the categories have their
own merit point. Using this method, student will be evaluated more
accurate compared to the previously. So, teacher can easily evaluate
their student without having any emotion factor, relation factor and
others. As conclusion this system module will helps the development
of student evaluation more accurate and valid in Student Idol.
Abstract: Learning programming is difficult for many learners. Some researches have found that the main difficulty relates to cognitive load. Cognitive overload happens in programming due to the nature of the subject which is intrinisicly over-bearing on the working memory. It happens due to the complexity of the subject itself. The problem is made worse by the poor instructional design methodology used in the teaching and learning process. Various efforts have been proposed to reduce the cognitive load, e.g. visualization softwares, part-program method etc. Use of many computer based systems have also been tried to tackle the problem. However, little success has been made to alleviate the problem. More has to be done to overcome this hurdle. This research attempts at understanding how cognitive load can be managed so as to reduce the problem of overloading. We propose a mechanism to measure the cognitive load during pre instruction, post instruction and in instructional stages of learning. This mechanism is used to help the instruction. As the load changes the instruction is made to adapt itself to ensure cognitive viability. This mechanism could be incorporated as a sub domain in the student model of various computer based instructional systems to facilitate the learning of programming.
Abstract: E-learning aims to build knowledge and skills in order
to enhance the quality of learning. Research has shown that the
majority of the e-learning solutions lack in pedagogical background
and present some serious deficiencies regarding teaching strategies
and content delivery, time and pace management, interface design
and preservation of learners- focus. The aim of this review is to
approach the design of e-learning solutions with a pedagogical
perspective and to present some good practices of e-learning design
grounded on the core principles of Learning Theories (LTs).
Abstract: Online discussions are an important component of
both blended and online courses. This paper examines the varieties of
online discussions and the perils, pitfalls and possibilities of this
rather new technological tool for enhanced learning. The discussion
begins with possible perils and pitfalls inherent in this educational
tool and moves to a consideration of the advantages of the varieties
of online discussions feasible for use in teacher education programs.
Abstract: Combining classifiers is a useful method for solving
complex problems in machine learning. The ECOC (Error Correcting
Output Codes) method has been widely used for designing combining
classifiers with an emphasis on the diversity of classifiers. In this
paper, in contrast to the standard ECOC approach in which individual
classifiers are chosen homogeneously, classifiers are selected
according to the complexity of the corresponding binary problem. We
use SATIMAGE database (containing 6 classes) for our experiments.
The recognition error rate in our proposed method is %10.37 which
indicates a considerable improvement in comparison with the
conventional ECOC and stack generalization methods.
Abstract: Modeling the behavior of the dialogue management in
the design of a spoken dialogue system using statistical methodologies
is currently a growing research area. This paper presents a work
on developing an adaptive learning approach to optimize dialogue
strategy. At the core of our system is a method formalizing dialogue
management as a sequential decision making under uncertainty whose
underlying probabilistic structure has a Markov Chain. Researchers
have mostly focused on model-free algorithms for automating the
design of dialogue management using machine learning techniques
such as reinforcement learning. But in model-free algorithms there
exist a dilemma in engaging the type of exploration versus exploitation.
Hence we present a model-based online policy learning
algorithm using interconnected learning automata for optimizing
dialogue strategy. The proposed algorithm is capable of deriving
an optimal policy that prescribes what action should be taken in
various states of conversation so as to maximize the expected total
reward to attain the goal and incorporates good exploration and
exploitation in its updates to improve the naturalness of humancomputer
interaction. We test the proposed approach using the most
sophisticated evaluation framework PARADISE for accessing to the
railway information system.
Abstract: Fuzzy Cognitive Maps (FCMs) have successfully
been applied in numerous domains to show relations between
essential components. In some FCM, there are more nodes, which
related to each other and more nodes means more complex in system
behaviors and analysis. In this paper, a novel learning method used to
construct FCMs based on historical data and by using data mining
and DEMATEL method, a new method defined to reduce nodes
number. This method cluster nodes in FCM based on their cause and
effect behaviors.
Abstract: Color image segmentation can be considered as a
cluster procedure in feature space. k-means and its adaptive
version, i.e. competitive learning approach are powerful tools
for data clustering. But k-means and competitive learning suffer
from several drawbacks such as dead-unit problem and need to
pre-specify number of cluster. In this paper, we will explore to
use competitive and cooperative learning approach to perform
color image segmentation. In competitive and cooperative
learning approach, seed points not only compete each other, but
also the winner will dynamically select several nearest
competitors to form a cooperative team to adapt to the input
together, finally it can automatically select the correct number
of cluster and avoid the dead-units problem. Experimental
results show that CCL can obtain better segmentation result.
Abstract: There have been many variations of technologies that helped educators in teaching & learning. From the past research it is evident that Information Technology significantly increases student participation and interactivity in the classrooms. This research started with a aim to find whether adoption of Wi-Fi environment by Malaysian Higher Educational Institutions (HEI) can benefit students and staff equally. The study was carried out in HEI-s of Klang Valley, Malaysia and the data is gathered through paper based surveys. A sample size of 237 units were randomly selected from 5 higher educational institutions in the Klang Valley using the Stratified Random sampling method and from the analysis of the data, it was found that the implementation of wireless technologies in HEIs have created lot of opportunities and also challenges.
Abstract: The development and use of mobile devices as well as its integration within education systems to deliver electronic contents and to support real-time communications was the focus of this research. In order to investigate the software engineering issues in using mobile devices a research on electronic content was initiated. The Developed MP3 mobile software solution was developed as a prototype for testing and developing a strategy for designing a usable m-learning environment. The mobile software solution was evaluated using mobile device using the link: http://projects.seeu.edu.mk/mlearn. The investigation also tested the correlation between the two mobile learning indicators: electronic content and attention, based on the Task Based learning instructional method. The mobile software solution ''M-Learn“ was developed as a prototype for testing the approach and developing a strategy for designing usable m-learning environment. The proposed methodology is about what learning modeling approach is more appropriate to use when developing mobile learning software.
Abstract: This paper presents a simple and effective method for approximate indexing of instances for instance based learning. The method uses an interval tree to determine a good starting search point for the nearest neighbor. The search stops when an early stopping criterion is met. The method proved to be very effective especially when only the first nearest neighbor is required.