Abstract: Classification of electroencephalogram (EEG) signals
extracted during mental tasks is a technique that is actively pursued
for Brain Computer Interfaces (BCI) designs. In this paper, we
compared the classification performances of univariateautoregressive
(AR) and multivariate autoregressive (MAR) models
for representing EEG signals that were extracted during different
mental tasks. Multilayer Perceptron (MLP) neural network (NN)
trained by the backpropagation (BP) algorithm was used to classify
these features into the different categories representing the mental
tasks. Classification performances were also compared across
different mental task combinations and 2 sets of hidden units (HU): 2
to 10 HU in steps of 2 and 20 to 100 HU in steps of 20. Five different
mental tasks from 4 subjects were used in the experimental study and
combinations of 2 different mental tasks were studied for each
subject. Three different feature extraction methods with 6th order
were used to extract features from these EEG signals: AR
coefficients computed with Burg-s algorithm (ARBG), AR
coefficients computed with stepwise least square algorithm (ARLS)
and MAR coefficients computed with stepwise least square
algorithm. The best results were obtained with 20 to 100 HU using
ARBG. It is concluded that i) it is important to choose the suitable
mental tasks for different individuals for a successful BCI design, ii)
higher HU are more suitable and iii) ARBG is the most suitable
feature extraction method.
Abstract: In this paper, we propose a new method to distinguish
between arousal and relaxation states by using multiple features
acquired from a photoplethysmogram (PPG) and support vector
machine (SVM). To induce arousal and relaxation states in subjects, 2
kinds of sound stimuli are used, and their corresponding biosignals are
obtained using the PPG sensor. Two features–pulse to pulse interval
(PPI) and pulse amplitude (PA)–are extracted from acquired PPG
data, and a nonlinear classification between arousal and relaxation is
performed using SVM.
This methodology has several advantages when compared with
previous similar studies. Firstly, we extracted 2 separate features from
PPG, i.e., PPI and PA. Secondly, in order to improve the classification
accuracy, SVM-based nonlinear classification was performed.
Thirdly, to solve classification problems caused by generalized
features of whole subjects, we defined each threshold according to
individual features.
Experimental results showed that the average classification
accuracy was 74.67%. Also, the proposed method showed the better
identification performance than the single feature based methods.
From this result, we confirmed that arousal and relaxation can be
classified using SVM and PPG features.
Abstract: Manufacturing companies are facing a broad variety
of challenges caused by a dynamic production environment. To
succeed in such an environment, it is crucial to minimize the loss of
time required to trigger the adaptation process of a company-s
production structures. This paper presents an approach for the
continuous monitoring of production structures by neurologic
principles. It enhances classical monitoring concepts, which are
principally focused on reactive strategies, and enables companies to
act proactively. Thereby, strategic aspects regarding the
harmonization of certain life cycles are integrated into the decision
making process for triggering the reconfiguration process of the
production structure.
Abstract: Power System Security is a major concern in real time
operation. Conventional method of security evaluation consists of
performing continuous load flow and transient stability studies by
simulation program. This is highly time consuming and infeasible
for on-line application. Pattern Recognition (PR) is a promising
tool for on-line security evaluation. This paper proposes a Support
Vector Machine (SVM) based binary classification for static and
transient security evaluation. The proposed SVM based PR approach
is implemented on New England 39 Bus and IEEE 57 Bus systems.
The simulation results of SVM classifier is compared with the other
classifier algorithms like Method of Least Squares (MLS), Multi-
Layer Perceptron (MLP) and Linear Discriminant Analysis (LDA)
classifiers.
Abstract: In order to be able to automatically differentiate
between two modes of permanent flow of a liquid simulating blood,
it was imperative to put together a data bank. Thus, the acquisition of
the various amplitude spectra of the Doppler signal of this liquid in
laminar flow and other spectra in turbulent flow enabled us to
establish an automatic difference between the two modes. According
to the number of parameters and their nature, a comparative study
allowed us to choose the best classifier.
Abstract: Concurrency and synchronization are becoming big
issues as every new PC comes with multi-core processors. A major
reason for Object-Oriented Programming originally was to enable
easier reuse: encode your algorithm into a class and thoroughly
debug it, then you can reuse the class again and again. However,
when we get to concurrency and synchronization, this is often not
possible. Thread-safety issues means that synchronization constructs
need to be entangled into every class involved. We contributed a
detailed literature review of issues and challenges in concurrent
programming and present a methodology that uses the Aspect-
Oriented paradigm to address this problem. Aspects will allow us to
extract the synchronization concerns as schemes to be “weaved in"
later into the main code. This allows the aspects to be separately
tested and verified. Hence, the functional components can be weaved
with reusable synchronization schemes that are robust and scalable.
Abstract: In this paper, numerical simulations are performed to investigate the effect of disturbance block on flow field of the classical square lid-driven cavity. Attentions are focused on vortex formation and studying the effect of block position on its structure. Corner vortices are different upon block position and new vortices are produced because of the block. Finite volume method is used to solve Navier-Stokes equations and PISO algorithm is employed for the linkage of velocity and pressure. Verification and grid independency of results are reported. Stream lines are sketched to visualize vortex structure in different block positions.
Abstract: Since the presentation of the backpropagation algorithm, a vast variety of improvements of the technique for training a feed forward neural networks have been proposed. This article focuses on two classes of acceleration techniques, one is known as Local Adaptive Techniques that are based on weightspecific only, such as the temporal behavior of the partial derivative of the current weight. The other, known as Dynamic Adaptation Methods, which dynamically adapts the momentum factors, α, and learning rate, η, with respect to the iteration number or gradient. Some of most popular learning algorithms are described. These techniques have been implemented and tested on several problems and measured in terms of gradient and error function evaluation, and percentage of success. Numerical evidence shows that these techniques improve the convergence of the Backpropagation algorithm.
Abstract: The problem of incompressible steady flow simulation around an airfoil is discussed. For some simplest airfoils (circular, elliptical, Zhukovsky airfoils) the exact solution is known from complex analysis. It allows to compute the intensity of vortex layer which simulates the airfoil. Some modifications of the vortex element method are proposed and test computations are carried out. It-s shown that the these approaches are much more effective in comparison with the classical numerical scheme.
Abstract: We present an Electronic Nose (ENose), which is
aimed at identifying the presence of one out of two gases, possibly
detecting the presence of a mixture of the two. Estimation of the
concentrations of the components is also performed for a volatile
organic compound (VOC) constituted by methanol and acetone, for
the ranges 40-400 and 22-220 ppm (parts-per-million), respectively.
Our system contains 8 sensors, 5 of them being gas sensors (of the
class TGS from FIGARO USA, INC., whose sensing element is a tin
dioxide (SnO2) semiconductor), the remaining being a temperature
sensor (LM35 from National Semiconductor Corporation), a
humidity sensor (HIH–3610 from Honeywell), and a pressure sensor
(XFAM from Fujikura Ltd.).
Our integrated hardware–software system uses some machine
learning principles and least square regression principle to identify at
first a new gas sample, or a mixture, and then to estimate the
concentrations. In particular we adopt a training model using the
Support Vector Machine (SVM) approach with linear kernel to teach
the system how discriminate among different gases. Then we apply
another training model using the least square regression, to predict
the concentrations.
The experimental results demonstrate that the proposed
multiclassification and regression scheme is effective in the
identification of the tested VOCs of methanol and acetone with
96.61% correctness. The concentration prediction is obtained with
0.979 and 0.964 correlation coefficient for the predicted versus real
concentrations of methanol and acetone, respectively.
Abstract: In this paper, we establish existence and uniqueness of
solutions for a class of inverse problems of degenerate differential
equations. The main tool is the perturbation theory for linear operators.
Abstract: An iterative algorithm is proposed and tested in Cournot Game models, which is based on the convergence of sequential best responses and the utilization of a genetic algorithm for determining each player-s best response to a given strategy profile of its opponents. An extra outer loop is used, to address the problem of finite accuracy, which is inherent in genetic algorithms, since the set of feasible values in such an algorithm is finite. The algorithm is tested in five Cournot models, three of which have convergent best replies sequence, one with divergent sequential best replies and one with “local NE traps"[14], where classical local search algorithms fail to identify the Nash Equilibrium. After a series of simulations, we conclude that the algorithm proposed converges to the Nash Equilibrium, with any level of accuracy needed, in all but the case where the sequential best replies process diverges.
Abstract: This paper presents the stabilization potential of Class
F pond ash (PA) from a coal fired thermal power station on tropical
peat soil. Peat or highly organic soils are well known for their high
compressibility, natural moisture content, low shear strength and
long-term settlement. This study investigates the effect of different
amount (i.e., 5, 10, 15 and 20%) of PA on peat soil, collected from
Sarawak, Malaysia, mainly compaction and unconfined compressive
strength (UCS) properties. The amounts of PA added to the peat soil
sample as percentage of the dry peat soil mass. With the increase in
PA content, the maximum dry density (MDD) of peat soil increases,
while the optimum moisture content (OMC) decreases. The UCS
value of the peat soils increases significantly with the increase of PA
content and also with curing periods. This improvement on
compressive strength of tropical peat soils indicates that PA has the
potential to be used as a stabilizer for tropical peat soil. Also, the use
of PA in soil stabilization helps in reducing the pond volume and
achieving environment friendly as well as a sustainable development
of natural resources.
Abstract: Software reuse can be considered as the most realistic
and promising way to improve software engineering productivity and
quality. Automated assistance for software reuse involves the
representation, classification, retrieval and adaptation of components.
The representation and retrieval of components are important to
software reuse in Component-Based on Software Development
(CBSD). However, current industrial component models mainly focus
on the implement techniques and ignore the semantic information
about component, so it is difficult to retrieve the components that
satisfy user-s requirements. This paper presents a method of business
component retrieval based on specification matching to solve the
software reuse of enterprise information system. First, a business
component model oriented reuse is proposed. In our model, the
business data type is represented as sign data type based on XML,
which can express the variable business data type that can describe the
variety of business operations. Based on this model, we propose
specification match relationships in two levels: business operation
level and business component level. In business operation level, we
use input business data types, output business data types and the
taxonomy of business operations evaluate the similarity between
business operations. In the business component level, we propose five
specification matches between business components. To retrieval
reusable business components, we propose the measure of similarity
degrees to calculate the similarities between business components.
Finally, a business component retrieval command like SQL is
proposed to help user to retrieve approximate business components
from component repository.
Abstract: In the fifteenth century, the Malacca Empire emerged
as the centre of Islamic civilization in the Malay Archipelago. The
history had been recorded in Sulalat Al-Salatin, an important literary
source about the genealogy of all Kings in Malacca. The objective of
this study was to analyze the understanding of sayings from Prophet
Muhammad among Malays in Malacca during the fifteenth century
through all of the hadith quoted in Sulalat Al-Salatin. This study used
content analysis methodology to validate the sayings where all of
them were critically analyzed and compared with the classical hadith
sources from prominent Muslim scholars. As a result, only two out of
the four quotations were considered as authentic sayings of Prophet
Muhammad. This study also showed the importance of the palace as
the centre of the Islamic education system and the role played by
Muslim preachers from outside of Malacca to propagate Islam in
Malacca.
Abstract: This paper presents a general trainable framework
for fast and robust upright human face and non-human object
detection and verification in static images. To enhance the
performance of the detection process, the technique we develop is
based on the combination of fast neural network (FNN) and
classical neural network (CNN). In FNN, a useful correlation is
exploited to sustain high level of detection accuracy between input
image and the weight of the hidden neurons. This is to enable the
use of Fourier transform that significantly speed up the time
detection. The combination of CNN is responsible to verify the
face region. A bootstrap algorithm is used to collect non human
object, which adds the false detection to the training process of the
human and non-human object. Experimental results on test images
with both simple and complex background demonstrate that the
proposed method has obtained high detection rate and low false
positive rate in detecting both human face and non-human object.
Abstract: Coloured Petri net (CPN) has been widely adopted in various areas in Computer Science, including protocol specification, performance evaluation, distributed systems and coordination in multi-agent systems. It provides a graphical representation of a system and has a strong mathematical foundation for proving various properties. This paper proposes a novel representation of a coloured Petri net using an extension of logic programming called abductive logic programming (ALP), which is purely based on classical logic. Under such a representation, an implementation of a CPN could be directly obtained, in which every inference step could be treated as a kind of equivalence preserved transformation. We would describe how to implement a CPN under such a representation using common meta-programming techniques in Prolog. We call our framework CPN-LP and illustrate its applications in modeling an intelligent agent.
Abstract: A new Markovianity approach is introduced in this
paper. This approach reduces the response time of classic Markov
Random Fields approach. First, one region is determinated by a
clustering technique. Then, this region is excluded from the study.
The remaining pixel form the study zone and they are selected for a
Markovianity segmentation task. With Selective Markovianity
approach, segmentation process is faster than classic one.
Abstract: In the present work, an attempt has been made to
understand the feasibility of using UHF technique for identification
of any corona discharges/ arcing in insulating material due to water
droplets. The sensors of broadband type are useful for identification
of such discharges. It is realised that arcing initiated by liquid droplet
radiates UHF signals in the entire bandwidth up to 2 GHz. The
frequency content of the UHF signal generated due to corona/arcing
is not much varied in epoxy nanocomposites with different weight
percentage of clay content. The exfoliated/intercalated properties
were analysed through TEM studies. It is realized that corona
initiated discharges are of intermittent process. The hydrophobicity
of the material characterized through contact angle measurement. It
is realized that low Wt % of nanoclay content in epoxy resin reduces
the surface carbonization due to arcing/corona discharges. The results
of the study with gamma irradiated specimen indicates that contact
angle, discharge inception time and evaporation time of the liquid are
much lower than the virgin epoxy nanocomposite material.
Abstract: KSLV-I(Korea Space Launch Vehicle-I) is designed as
a launch vehicle to enter a 100 kg-class satellite to the LEO(Low Earth
Orbit). Attitude angles of the upper-stage, including roll, pitch and
yaw are controlled by the cold gas thruster system using nitrogen gas.
The cold gas thruster is an actuator in the RCS(Reaction Control
System). To design an attitude controller for the upper-stage, thrust
measurement in vacuum condition is required. In this paper, the new
thrust measurement system and calibration mechanism are developed
and measurement errors and signal processing method are presented.