Abstract: Text categorization is the problem of classifying text
documents into a set of predefined classes. In this paper, we
investigated three approaches to build a meta-classifier in order to
increase the classification accuracy. The basic idea is to learn a metaclassifier
to optimally select the best component classifier for each
data point. The experimental results show that combining classifiers
can significantly improve the accuracy of classification and that our
meta-classification strategy gives better results than each individual
classifier. For 7083 Reuters text documents we obtained a
classification accuracies up to 92.04%.
Abstract: The issue of unintentional islanding in PV grid
interconnection still remains as a challenge in grid-connected
photovoltaic (PV) systems. This paper discusses the overview of
popularly used anti-islanding detection methods, practically applied
in PV grid-connected systems. Anti-islanding methods generally can
be classified into four major groups, which include passive methods,
active methods, hybrid methods and communication base methods.
Active methods have been the preferred detection technique over the
years due to very small non-detected zone (NDZ) in small scale
distribution generation. Passive method is comparatively simpler
than active method in terms of circuitry and operations. However, it
suffers from large NDZ that significantly reduces its performance.
Communication base methods inherit the advantages of active and
passive methods with reduced drawbacks. Hybrid method which
evolved from the combination of both active and passive methods
has been proven to achieve accurate anti-islanding detection by many
researchers. For each of the studied anti-islanding methods, the
operation analysis is described while the advantages and
disadvantages are compared and discussed. It is difficult to pinpoint a
generic method for a specific application, because most of the
methods discussed are governed by the nature of application and
system dependent elements. This study concludes that the setup and
operation cost is the vital factor for anti-islanding method selection in
order to achieve minimal compromising between cost and system
quality.
Abstract: This paper proposes a method, combining color and
layout features, for identifying documents captured from lowresolution
handheld devices. On one hand, the document image color
density surface is estimated and represented with an equivalent
ellipse and on the other hand, the document shallow layout structure
is computed and hierarchically represented. The combined color and
layout features are arranged in a symbolic file, which is unique for
each document and is called the document-s visual signature. Our
identification method first uses the color information in the
signatures in order to focus the search space on documents having a
similar color distribution, and finally selects the document having the
most similar layout structure in the remaining search space. Finally,
our experiment considers slide documents, which are often captured
using handheld devices.
Abstract: This paper presents a simplified version of Data Envelopment Analysis (DEA) - a conventional approach to evaluating the performance and ranking of competitive objects characterized by two groups of factors acting in opposite directions: inputs and outputs. DEA with a Perfect Object (DEA PO) augments the group of actual objects with a virtual Perfect Object - the one having greatest outputs and smallest inputs. It allows for obtaining an explicit analytical solution and making a step to an absolute efficiency. This paper develops this approach further and introduces a DEA model with Partially Perfect Objects. DEA PPO consecutively eliminates the smallest relative inputs or greatest relative outputs, and applies DEA PO to the reduced collections of indicators. The partial efficiency scores are combined to get the weighted efficiency score. The computational scheme remains simple, like that of DEA PO, but the advantage of the DEA PPO is taking into account all of the inputs and outputs for each actual object. Firm evaluation is considered as an example.
Abstract: A conventional binding method for low power in a
high-level synthesis mainly focuses on finding an optimal binding for
an assumed input data, and obtains only one binding table. In this
paper, we show that a binding method which uses multiple binding
tables gets better solution compared with the conventional methods
which use a single binding table, and propose a dynamic bus binding
scheme for low power using multiple binding tables. The proposed
method finds multiple binding tables for the proper partitions of an
input data, and switches binding tables dynamically to produce the
minimum total switching activity. Experimental result shows that the
proposed method obtains a binding solution having 12.6-28.9%
smaller total switching activity compared with the conventional
methods.
Abstract: Blood pulse is an important human physiological signal commonly used for the understanding of the individual physical health. Current methods of non-invasive blood pulse sensing require direct contact or access to the human skin. As such, the performances of these devices tend to vary with time and are subjective to human body fluids (e.g. blood, perspiration and skin-oil) and environmental contaminants (e.g. mud, water, etc). This paper proposes a simulation model for the novel method of non-invasive acquisition of blood pulse using the disturbance created by blood flowing through a localized magnetic field. The simulation model geometry represents a blood vessel, a permanent magnet, a magnetic sensor, surrounding tissues and air in 2-dimensional. In this model, the velocity and pressure fields in the blood stream are described based on Navier-Stroke equations and the walls of the blood vessel are assumed to have no-slip condition. The blood assumes a parabolic profile considering a laminar flow for blood in major artery near the skin. And the inlet velocity follows a sinusoidal equation. This will allow the computational software to compute the interactions between the magnetic vector potential generated by the permanent magnet and the magnetic nanoparticles in the blood. These interactions are simulated based on Maxwell equations at the location where the magnetic sensor is placed. The simulated magnetic field at the sensor location is found to assume similar sinusoidal waveform characteristics as the inlet velocity of the blood. The amplitude of the simulated waveforms at the sensor location are compared with physical measurements on human subjects and found to be highly correlated.
Abstract: Statistical learning theory was developed by Vapnik. It
is a learning theory based on Vapnik-Chervonenkis dimension. It also
has been used in learning models as good analytical tools. In general, a
learning theory has had several problems. Some of them are local
optima and over-fitting problems. As well, statistical learning theory
has same problems because the kernel type, kernel parameters, and
regularization constant C are determined subjectively by the art of
researchers. So, we propose an evolutionary statistical learning theory
to settle the problems of original statistical learning theory.
Combining evolutionary computing into statistical learning theory,
our theory is constructed. We verify improved performances of an
evolutionary statistical learning theory using data sets from KDD cup.
Abstract: An optical fiber Fabry-Perot interferometer (FFPI) is
proposed and demonstrated for dynamic measurements in a
mechanical vibrating target. A polishing metal with a low reflectance
value adhered to a mechanical vibrator was excited via a function
generator at various excitation frequencies. Output interference
fringes were generated by modulating the reference and sensing
signal at the output arm. A fringe-counting technique was used for
interpreting the displacement information on the dedicated computer.
The fiber interferometer has been found the capability of the
displacement measurements of 1.28 μm – 96.01 μm. A commercial
displacement sensor was employed as a reference sensor for
investigating the measurement errors from the fiber sensor. A
maximum percentage measurement error of approximately 1.59 %
was obtained.
Abstract: This paper introduces a temporal epistemic logic
CBCTL that updates agent-s belief states through communications
in them, based on computational tree logic (CTL). In practical
environments, communication channels between agents may not be
secure, and in bad cases agents might suffer blackouts. In this study,
we provide inform* protocol based on ACL of FIPA, and declare the
presence of secure channels between two agents, dependent on time.
Thus, the belief state of each agent is updated along with the progress
of time. We show a prover, that is a reasoning system for a given
formula in a given a situation of an agent ; if it is directly provable
or if it could be validated through the chains of communications, the
system returns the proof.
Abstract: The design of a modern aircraft is based on three pillars: theoretical results, experimental test and computational simulations.
As a results of this, Computational Fluid Dynamic (CFD) solvers are
widely used in the aeronautical field. These solvers require the correct
selection of many parameters in order to obtain successful results. Besides, the computational time spent in the simulation depends on
the proper choice of these parameters.
In this paper we create an expert system capable of making an
accurate prediction of the number of iterations and time required for the convergence of a computational fluid dynamic (CFD) solver.
Artificial neural network (ANN) has been used to design the expert system. It is shown that the developed expert system is capable of making an accurate prediction the number of iterations and time
required for the convergence of a CFD solver.
Abstract: Software estimation accuracy is among the greatest
challenges for software developers. This study aimed at building and
evaluating a neuro-fuzzy model to estimate software projects
development time. The forty-one modules developed from ten
programs were used as dataset. Our proposed approach is compared
with fuzzy logic and neural network model and Results show that the
value of MMRE (Mean of Magnitude of Relative Error) applying
neuro-fuzzy was substantially lower than MMRE applying fuzzy
logic and neural network.
Abstract: The authors present an algorithm for order reduction of linear dynamic systems using the combined advantages of stability equation method and the error minimization by Genetic algorithm. The denominator of the reduced order model is obtained by the stability equation method and the numerator terms of the lower order transfer function are determined by minimizing the integral square error between the transient responses of original and reduced order models using Genetic algorithm. The reduction procedure is simple and computer oriented. It is shown that the algorithm has several advantages, e.g. the reduced order models retain the steady-state value and stability of the original system. The proposed algorithm has also been extended for the order reduction of linear multivariable systems. Two numerical examples are solved to illustrate the superiority of the algorithm over some existing ones including one example of multivariable system.
Abstract: Conventionally the selection of parameters depends
intensely on the operator-s experience or conservative technological
data provided by the EDM equipment manufacturers that assign
inconsistent machining performance. The parameter settings given by
the manufacturers are only relevant with common steel grades. A
single parameter change influences the process in a complex way.
Hence, the present research proposes artificial neural network (ANN)
models for the prediction of surface roughness on first commenced
Ti-15-3 alloy in electrical discharge machining (EDM) process. The
proposed models use peak current, pulse on time, pulse off time and
servo voltage as input parameters. Multilayer perceptron (MLP) with
three hidden layer feedforward networks are applied. An assessment
is carried out with the models of distinct hidden layer. Training of the
models is performed with data from an extensive series of
experiments utilizing copper electrode as positive polarity. The
predictions based on the above developed models have been verified
with another set of experiments and are found to be in good
agreement with the experimental results. Beside this they can be
exercised as precious tools for the process planning for EDM.
Abstract: The paper proposes an approach using genetic algorithm for computing the region based image similarity. The image is denoted using a set of segmented regions reflecting color and texture properties of an image. An image is associated with a family of image features corresponding to the regions. The resemblance of two images is then defined as the overall similarity between two families of image features, and quantified by a similarity measure, which integrates properties of all the regions in the images. A genetic algorithm is applied to decide the most plausible matching. The performance of the proposed method is illustrated using examples from an image database of general-purpose images, and is shown to produce good results.
Abstract: In this paper we compare the response of linear and
nonlinear neural network-based prediction schemes in prediction of
received Signal-to-Interference Power Ratio (SIR) in Direct
Sequence Code Division Multiple Access (DS/CDMA) systems. The
nonlinear predictor is Multilayer Perceptron MLP and the linear
predictor is an Adaptive Linear (Adaline) predictor. We solve the
problem of complexity by using the Minimum Mean Squared Error
(MMSE) principle to select the optimal predictors. The optimized
Adaline predictor is compared to optimized MLP by employing
noisy Rayleigh fading signals with 1.8 GHZ carrier frequency in an
urban environment. The results show that the Adaline predictor can
estimates SIR with the same error as MLP when the user has the
velocity of 5 km/h and 60 km/h but by increasing the velocity up-to
120 km/h the mean squared error of MLP is two times more than
Adaline predictor. This makes the Adaline predictor (with lower
complexity) more suitable than MLP for closed-loop power control
where efficient and accurate identification of the time-varying
inverse dynamics of the multi path fading channel is required.
Abstract: Supplier selection is a multi criteria decision-making process that comprises tangible and intangible factors. The majority of previous supplier selection techniques do not consider strategic perspective. Besides, uncertainty is one of the most important obstacles in supplier selection. For the first, time in this paper, the idea of the algorithm " Knapsack " is used to select suppliers Moreover, an attempt has to be made to take the advantage of a simple numerical method for solving model .This is an innovation to resolve any ambiguity in choosing suppliers. This model has been tried in the suppliers selected in a competitive environment and according to all desired standards of quality and quantity to show the efficiency of the model, an industry sample has been uses.
Abstract: Reconfigurable optical add/drop multiplexers
(ROADMs) can be classified into three categories based on their
underlying switching technologies. Category I consists of a single
large optical switch; category II is composed of a number of small
optical switches aligned in parallel; and category III has a single
optical switch and only one wavelength being added/dropped. In this
paper, to evaluate the wavelength-routing capability of ROADMs of
category-II in dynamic optical networks,the dynamic traffic models
are designed based on Bernoulli, Poisson distributions for smooth
and regular types of traffic. Through Analytical and Simulation
results, the routing power of cat-II of ROADM networks for two
traffic models are determined.
Abstract: Through the time, the higher education has changed
the learning system since mother tongue to bilingual, and in this new
century has been coming develop a multilingual education. All as
part of globalization process of the countries and the education.
Nevertheless, this change only has been effectively in countries of the
first world, the rest have been lagging. Therefore, these countries
require strengthen their higher education systems through models that
give way to multilingual and bilingual education. In this way, shows
a new model adapted from a systemic form to allow a higher
bilingual and multilingual education in Latin America. This
systematization aims to increase the skills and competencies
student’s, decrease the time learning of a second tongue, add to
multilingualism in the American Latin Universities, also, contribute
to position the region´s countries in a better global status, and
stimulate the development of new research in this area.
Abstract: Different numerical methods are employed and developed for simulating interfacial flows. A large range of applications belong to this group, e.g. two-phase flows of air bubbles in water or water drops in air. In such problems surface tension effects often play a dominant role. In this paper, various models of surface tension force for interfacial flows, the CSF, CSS, PCIL and SGIP models have been applied to simulate the motion of small air bubbles in water and the results were compared and reviewed. It has been pointed out that by using SGIP or PCIL models, we are able to simulate bubble rise and obtain results in close agreement with the experimental data.
Abstract: Heterogeneity of solid waste characteristics as well as the complex processes taking place within the landfill ecosystem motivated the implementation of soft computing methodologies such as artificial neural networks (ANN), fuzzy logic (FL), and their combination. The present work uses a hybrid ANN-FL model that employs knowledge-based FL to describe the process qualitatively and implements the learning algorithm of ANN to optimize model parameters. The model was developed to simulate and predict the landfill gas production at a given time based on operational parameters. The experimental data used were compiled from lab-scale experiment that involved various operating scenarios. The developed model was validated and statistically analyzed using F-test, linear regression between actual and predicted data, and mean squared error measures. Overall, the simulated landfill gas production rates demonstrated reasonable agreement with actual data. The discussion focused on the effect of the size of training datasets and number of training epochs.