Abstract: This paper presents a new approach for the prob-ability density function estimation using the Support Vector Ma-chines (SVM) and the Expectation Maximization (EM) algorithms.In the proposed approach, an advanced algorithm for the SVM den-sity estimation which incorporates the Mean Field theory in the learning process is used. Instead of using ad-hoc values for the para-meters of the kernel function which is used by the SVM algorithm,the proposed approach uses the EM algorithm for an automatic optimization of the kernel. Experimental evaluation using simulated data set shows encouraging results.
Abstract: This paper explores the scalability issues associated
with solving the Named Entity Recognition (NER) problem using
Support Vector Machines (SVM) and high-dimensional features. The
performance results of a set of experiments conducted using binary
and multi-class SVM with increasing training data sizes are
examined. The NER domain chosen for these experiments is the
biomedical publications domain, especially selected due to its
importance and inherent challenges. A simple machine learning
approach is used that eliminates prior language knowledge such as
part-of-speech or noun phrase tagging thereby allowing for its
applicability across languages. No domain-specific knowledge is
included. The accuracy measures achieved are comparable to those
obtained using more complex approaches, which constitutes a
motivation to investigate ways to improve the scalability of multiclass
SVM in order to make the solution more practical and useable.
Improving training time of multi-class SVM would make support
vector machines a more viable and practical machine learning
solution for real-world problems with large datasets. An initial
prototype results in great improvement of the training time at the
expense of memory requirements.
Abstract: This paper presents a new approach for the protection
of Thyristor-Controlled Series Compensator (TCSC) line using
Support Vector Machine (SVM). One SVM is trained for fault
classification and another for section identification. This method use
three phase current measurement that results in better speed and
accuracy than other SVM based methods which used single phase
current measurement. This makes it suitable for real-time protection.
The method was tested on 10,000 data instances with a very wide
variation in system conditions such as compensation level, source
impedance, location of fault, fault inception angle, load angle at
source bus and fault resistance. The proposed method requires only
local current measurement.
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.
Abstract: The winding hot-spot temperature is one of the most
critical parameters that affect the useful life of the power
transformers. The winding hot-spot temperature can be calculated as
function of the top-oil temperature that can estimated by using the
ambient temperature and transformer loading measured data. This
paper proposes the estimation of the top-oil temperature by using a
method based on Least Squares Support Vector Machines approach.
The estimated top-oil temperature is compared with measured data of
a power transformer in operation. The results are also compared with
methods based on the IEEE Standard C57.91-1995/2000 and
Artificial Neural Networks. It is shown that the Least Squares
Support Vector Machines approach presents better performance than
the methods based in the IEEE Standard C57.91-1995/2000 and
artificial neural networks.
Abstract: The interaction between wakes of bluff body and
airfoil have profound influences on system performance in many
industrial applications, e.g., turbo-machinery and cooling fan. The
present work investigates the effect of configuration include; airfoil-s
angle of attack, transverse and inline spacing of the models, on
frequency behavior of the cylinder-s near-wake. The experiments
carried on under subcritical flow regime, using the hot-wire
anemometry (HWA). The relationship between the Strouhal numbers
and arrangements provide an insight into the global physical
processes of wake interaction and vortex shedding.
Abstract: This paper represents four unsupervised clustering algorithms namely sIB, RandomFlatClustering, FarthestFirst, and FilteredClusterer that previously works have not been used for network traffic classification. The methodology, the result, the products of the cluster and evaluation of these algorithms with efficiency of each algorithm from accuracy are shown. Otherwise, the efficiency of these algorithms considering form the time that it use to generate the cluster quickly and correctly. Our work study and test the best algorithm by using classify traffic anomaly in network traffic with different attribute that have not been used before. We analyses the algorithm that have the best efficiency or the best learning and compare it to the previously used (K-Means). Our research will be use to develop anomaly detection system to more efficiency and more require in the future.
Abstract: Mathematical and computational modeling of calcium
signalling in nerve cells has produced considerable insights into how
the cells contracts with other cells under the variation of biophysical
and physiological parameters. The modeling of calcium signaling in
astrocytes has become more sophisticated. The modeling effort has
provided insight to understand the cell contraction. Main objective
of this work is to study the effect of voltage gated (Operated)
calcium channel (VOC) on calcium profile in the form of advection
diffusion equation. A mathematical model is developed in the form
of advection diffusion equation for the calcium profile. The model
incorporates the important physiological parameter like diffusion
coefficient etc. Appropriate boundary conditions have been framed.
Finite volume method is employed to solve the problem. A program
has been developed using in MATLAB 7.5 for the entire problem
and simulated on an AMD-Turion 32-bite machine to compute the
numerical results.
Abstract: Recent years have witnessed the rapid development of
the Internet and telecommunication techniques. Information security
is becoming more and more important. Applications such as covert
communication, copyright protection, etc, stimulate the research of
information hiding techniques. Traditionally, encryption is used to
realize the communication security. However, important information
is not protected once decoded. Steganography is the art and science
of communicating in a way which hides the existence of the communication.
Important information is firstly hidden in a host data, such
as digital image, video or audio, etc, and then transmitted secretly
to the receiver.In this paper a data hiding model with high security
features combining both cryptography using finite state sequential
machine and image based steganography technique for communicating
information more securely between two locations is proposed.
The authors incorporated the idea of secret key for authentication
at both ends in order to achieve high level of security. Before the
embedding operation the secret information has been encrypted with
the help of finite-state sequential machine and segmented in different
parts. The cover image is also segmented in different objects through
normalized cut.Each part of the encoded secret information has been
embedded with the help of a novel image steganographic method
(PMM) on different cuts of the cover image to form different stego
objects. Finally stego image is formed by combining different stego
objects and transmit to the receiver side. At the receiving end different
opposite processes should run to get the back the original secret
message.
Abstract: This paper proposes a direct power control for
doubly-fed induction machine for variable speed wind power
generation. It provides decoupled regulation of the primary side
active and reactive power and it is suitable for both electric energy
generation and drive applications. In order to control the power
flowing between the stator of the DFIG and the network, a decoupled
control of active and reactive power is synthesized using PI
controllers.The obtained simulation results show the feasibility
and the effectiveness of the suggested method
Abstract: Complex assemblies of interacting proteins carry out
most of the interesting jobs in a cell, such as metabolism, DNA
synthesis, mitosis and cell division. These physiological properties
play out as a subtle molecular dance, choreographed by underlying
regulatory networks that control the activities of cyclin-dependent
kinases (CDK). The network can be modeled by a set of nonlinear
differential equations and its behavior predicted by numerical
simulation. In this paper, an innovative approach has been proposed
that uses genetic algorithms to mine a set of behavior data output by
a biological system in order to determine the kinetic parameters of
the system. In our approach, the machine learning method is
integrated with the framework of existent biological information in a
wiring diagram so that its findings are expressed in a form of system
dynamic behavior. By numerical simulations it has been illustrated
that the model is consistent with experiments and successfully shown
that such application of genetic algorithms will highly improve the
performance of mathematical model of the cell division cycle to
simulate such a complicated bio-system.
Abstract: Human perceives color in categories, which may be
identified using color name such as red, blue, etc. The categorization
is unique for each human being. However despite the individual
differences, the categorization is shared among members in society.
This allows communication among them, especially when using
color name. Sociable robot, to live coexist with human and become
part of human society, must also have the shared color
categorization, which can be achieved through learning. Many
works have been done to enable computer, as brain of robot, to learn
color categorization. Most of them rely on modeling of human color
perception and mathematical complexities. Differently, in this work,
the computer learns color categorization through interaction with
humans. This work aims at developing the innate ability of the
computer to learn the human-like color categorization. It focuses on
the representation of color categorization and how it is built and
developed without much mathematical complexity.
Abstract: Chaiyaphum Starch Co. Ltd. is one of many starch
manufacturers that has introduced machinery to aid in manufacturing.
Even though machinery has replaced many elements and is now a
significant part in manufacturing processes, problems that must be
solved with respect to current process flow to increase efficiency still
exist. The paper-s aim is to increase productivity while maintaining
desired quality of starch, by redesigning the flipping machine-s
mechanical control system which has grossly low functional lifetime.
Such problems stem from the mechanical control system-s bearings,
as fluids and humidity can access into said bearing directly, in
tandem with vibrations from the machine-s function itself. The wheel
which is used to sense starch thickness occasionally falls from its
shaft, due to high speed rotation during operation, while the shaft
may bend from impact when processing dried bread. Redesigning its
mechanical control system has increased its efficiency, allowing
quality thickness measurement while increasing functional lifetime
an additional 62 days.
Abstract: Given a parallel program to be executed on a heterogeneous
computing system, the overall execution time of the program
is determined by a schedule. In this paper, we analyze the worst-case
performance of the list scheduling algorithm for scheduling tasks
of a parallel program in a mixed-machine heterogeneous computing
system such that the total execution time of the program is minimized.
We prove tight lower and upper bounds for the worst-case
performance ratio of the list scheduling algorithm. We also examine
the average-case performance of the list scheduling algorithm. Our
experimental data reveal that the average-case performance of the list
scheduling algorithm is much better than the worst-case performance
and is very close to optimal, except for large systems with large
heterogeneity. Thus, the list scheduling algorithm is very useful in
real applications.
Abstract: Tumor classification is a key area of research in the
field of bioinformatics. Microarray technology is commonly used in
the study of disease diagnosis using gene expression levels. The
main drawback of gene expression data is that it contains thousands
of genes and a very few samples. Feature selection methods are used
to select the informative genes from the microarray. These methods
considerably improve the classification accuracy. In the proposed
method, Genetic Algorithm (GA) is used for effective feature
selection. Informative genes are identified based on the T-Statistics,
Signal-to-Noise Ratio (SNR) and F-Test values. The initial candidate
solutions of GA are obtained from top-m informative genes. The
classification accuracy of k-Nearest Neighbor (kNN) method is used
as the fitness function for GA. In this work, kNN and Support Vector
Machine (SVM) are used as the classifiers. The experimental results
show that the proposed work is suitable for effective feature
selection. With the help of the selected genes, GA-kNN method
achieves 100% accuracy in 4 datasets and GA-SVM method
achieves in 5 out of 10 datasets. The GA with kNN and SVM
methods are demonstrated to be an accurate method for microarray
based tumor classification.
Abstract: It has become crucial over the years for nations to
improve their credit scoring methods and techniques in light of the
increasing volatility of the global economy. Statistical methods or
tools have been the favoured means for this; however artificial
intelligence or soft computing based techniques are becoming
increasingly preferred due to their proficient and precise nature and
relative simplicity. This work presents a comparison between Support
Vector Machines and Artificial Neural Networks two popular soft
computing models when applied to credit scoring. Amidst the
different criteria-s that can be used for comparisons; accuracy,
computational complexity and processing times are the selected
criteria used to evaluate both models. Furthermore the German credit
scoring dataset which is a real world dataset is used to train and test
both developed models. Experimental results obtained from our study
suggest that although both soft computing models could be used with
a high degree of accuracy, Artificial Neural Networks deliver better
results than Support Vector Machines.
Abstract: Several combinations of the preprocessing algorithms,
feature selection techniques and classifiers can be applied to the data
classification tasks. This study introduces a new accurate classifier,
the proposed classifier consist from four components: Signal-to-
Noise as a feature selection technique, support vector machine,
Bayesian neural network and AdaBoost as an ensemble algorithm.
To verify the effectiveness of the proposed classifier, seven well
known classifiers are applied to four datasets. The experiments show
that using the suggested classifier enhances the classification rates for
all datasets.
Abstract: This paper investigates the occurrence of regenerative
chatter vibrations in facing and turning processes. Orthogonal turning
(facing) and normal turning experiments are carried out under stable
as well as in the presence of controlled chatter vibrations. The effects
of chatter vibrations on various sensor signals are captured and
analyzed using frequency domain methods, which successfully
detected the chatter vibrations close to the dominant mode of the
machine tool system.
Abstract: In the last years, the computers have increased their capacity of calculus and networks, for the interconnection of these machines. The networks have been improved until obtaining the actual high rates of data transferring. The programs that nowadays try to take advantage of these new technologies cannot be written using the traditional techniques of programming, since most of the algorithms were designed for being executed in an only processor,in a nonconcurrent form instead of being executed concurrently ina set of processors working and communicating through a network.This paper aims to present the ongoing development of a new system for the reconfiguration of grouping of computers, taking into account these new technologies.
Abstract: This paper presents dynamic voltage collapse prediction on an actual power system using support vector machines.
Dynamic voltage collapse prediction is first determined based on the PTSI calculated from information in dynamic simulation output. Simulations were carried out on a practical 87 bus test system by considering load increase as the contingency. The data collected from the time domain simulation is then used as input to the SVM in which support vector regression is used as a predictor to determine the
dynamic voltage collapse indices of the power system. To reduce training time and improve accuracy of the SVM, the Kernel function type and Kernel parameter are considered. To verify the
effectiveness of the proposed SVM method, its performance is compared with the multi layer perceptron neural network (MLPNN). Studies show that the SVM gives faster and more accurate results for dynamic voltage collapse prediction compared with the MLPNN.