Abstract: In this work, a training algorithm for probabilistic neural networks (PNN) is presented. The algorithm addresses one of the major drawbacks of PNN, which is the size of the hidden layer in the network. By using a cross-validation training algorithm, the number of hidden neurons is shrunk to a smaller number consisting of the most representative samples of the training set. This is done without affecting the overall architecture of the network. Performance of the network is compared against performance of standard PNN for different databases from the UCI database repository. Results show an important gain in network size and performance.
Abstract: This paper has critically examined the use of Machine Learning procedures in curbing unauthorized access into valuable areas of an organization. The use of passwords, pin codes, user’s identification in recent times has been partially successful in curbing crimes involving identities, hence the need for the design of a system which incorporates biometric characteristics such as DNA and pattern recognition of variations in facial expressions. The facial model used is the OpenCV library which is based on the use of certain physiological features, the Raspberry Pi 3 module is used to compile the OpenCV library, which extracts and stores the detected faces into the datasets directory through the use of camera. The model is trained with 50 epoch run in the database and recognized by the Local Binary Pattern Histogram (LBPH) recognizer contained in the OpenCV. The training algorithm used by the neural network is back propagation coded using python algorithmic language with 200 epoch runs to identify specific resemblance in the exclusive OR (XOR) output neurons. The research however confirmed that physiological parameters are better effective measures to curb crimes relating to identities.
Abstract: Nowadays, tunnels with different applications are developed, and most of them are related to subway tunnels. The excavation of shallow tunnels that pass under municipal utilities is very important, and the surface settlement control is an important factor in the design. The study sought to analyze the settlement and also to find an appropriate model in order to predict the behavior of the tunnel in Tehran subway line-3. The displacement in these sections is also determined by using numerical analyses and numerical modeling. In addition, the Adaptive Neuro-Fuzzy Inference System (ANFIS) method is utilized by Hybrid training algorithm. The database pertinent to the optimum network was obtained from 46 subway tunnels in Iran and Turkey which have been constructed by the new Austrian tunneling method (NATM) with similar parameters based on type of their soil. The surface settlement was measured, and the acquired results were compared to the predicted values. The results disclosed that computing intelligence is a good substitute for numerical modeling.
Abstract: Mumbai, being traditionally the epicenter of India's
trade and commerce, the existing major ports such as Mumbai and
Jawaharlal Nehru Ports (JN) situated in Thane estuary are also
developing its waterfront facilities. Various developments over the
passage of decades in this region have changed the tidal flux
entering/leaving the estuary. The intake at Pir-Pau is facing the
problem of shortage of water in view of advancement of shoreline,
while jetty near Ulwe faces the problem of ship scheduling due to
existence of shallower depths between JN Port and Ulwe Bunder. In
order to solve these problems, it is inevitable to have information
about tide levels over a long duration by field measurements.
However, field measurement is a tedious and costly affair;
application of artificial intelligence was used to predict water levels
by training the network for the measured tide data for one lunar tidal
cycle. The application of two layered feed forward Artificial Neural
Network (ANN) with back-propagation training algorithms such as
Gradient Descent (GD) and Levenberg-Marquardt (LM) was used to
predict the yearly tide levels at waterfront structures namely at Ulwe
Bunder and Pir-Pau. The tide data collected at Apollo Bunder, Ulwe,
and Vashi for a period of lunar tidal cycle (2013) was used to train,
validate and test the neural networks. These trained networks having
high co-relation coefficients (R= 0.998) were used to predict the tide
at Ulwe, and Vashi for its verification with the measured tide for the
year 2000 & 2013. The results indicate that the predicted tide levels
by ANN give reasonably accurate estimation of tide. Hence, the
trained network is used to predict the yearly tide data (2015) for
Ulwe. Subsequently, the yearly tide data (2015) at Pir-Pau was
predicted by using the neural network which was trained with the
help of measured tide data (2000) of Apollo and Pir-Pau. The analysis of measured data and study reveals that: The
measured tidal data at Pir-Pau, Vashi and Ulwe indicate that there is
maximum amplification of tide by about 10-20 cm with a phase lag
of 10-20 minutes with reference to the tide at Apollo Bunder
(Mumbai). LM training algorithm is faster than GD and with increase
in number of neurons in hidden layer and the performance of the
network increases. The predicted tide levels by ANN at Pir-Pau and
Ulwe provides valuable information about the occurrence of high and
low water levels to plan the operation of pumping at Pir-Pau and
improve ship schedule at Ulwe.
Abstract: Brain-Computer Interfaces (BCIs) measure brain
signals activity, intentionally and unintentionally induced by users,
and provides a communication channel without depending on the
brain’s normal peripheral nerves and muscles output pathway.
Feature Selection (FS) is a global optimization machine learning
problem that reduces features, removes irrelevant and noisy data
resulting in acceptable recognition accuracy. It is a vital step
affecting pattern recognition system performance. This study presents
a new Binary Particle Swarm Optimization (BPSO) based feature
selection algorithm. Multi-layer Perceptron Neural Network
(MLPNN) classifier with backpropagation training algorithm and
Levenberg-Marquardt training algorithm classify selected features.
Abstract: Margin-Based Principle has been proposed for a long
time, it has been proved that this principle could reduce the
structural risk and improve the performance in both theoretical
and practical aspects. Meanwhile, feed-forward neural network is
a traditional classifier, which is very hot at present with a deeper
architecture. However, the training algorithm of feed-forward neural
network is developed and generated from Widrow-Hoff Principle that
means to minimize the squared error. In this paper, we propose
a new training algorithm for feed-forward neural networks based
on Margin-Based Principle, which could effectively promote the
accuracy and generalization ability of neural network classifiers
with less labelled samples and flexible network. We have conducted
experiments on four UCI open datasets and achieved good results
as expected. In conclusion, our model could handle more sparse
labelled and more high-dimension dataset in a high accuracy while
modification from old ANN method to our method is easy and almost
free of work.
Abstract: Artificial Neural Network (ANN) can be trained using
back propagation (BP). It is the most widely used algorithm for
supervised learning with multi-layered feed-forward networks.
Efficient learning by the BP algorithm is required for many practical
applications. The BP algorithm calculates the weight changes of
artificial neural networks, and a common approach is to use a twoterm
algorithm consisting of a learning rate (LR) and a momentum
factor (MF). The major drawbacks of the two-term BP learning
algorithm are the problems of local minima and slow convergence
speeds, which limit the scope for real-time applications. Recently the
addition of an extra term, called a proportional factor (PF), to the
two-term BP algorithm was proposed. The third increases the speed
of the BP algorithm. However, the PF term also reduces the
convergence of the BP algorithm, and criteria for evaluating
convergence are required to facilitate the application of the three
terms BP algorithm. Although these two seem to be closely related,
as described later, we summarize various improvements to overcome
the drawbacks. Here we compare the different methods of
convergence of the new three-term BP algorithm.
Abstract: In this research, the capability of neural networks in
modeling and learning complicated and nonlinear relations has been
used to develop a model for the prediction of changes in the diameter
of bubbles in pool boiling distilled water. The input parameters used
in the development of this network include element temperature, heat
flux, and retention time of bubbles. The test data obtained from the
experiment of the pool boiling of distilled water, and the
measurement of the bubbles form on the cylindrical element. The
model was developed based on training algorithm, which is
typologically of back-propagation type. Considering the correlation
coefficient obtained from this model is 0.9633. This shows that this
model can be trusted for the simulation and modeling of the size of
bubble and thermal transfer of boiling.
Abstract: This paper proposes a neural network weights and
topology optimization using genetic evolution and the
backpropagation training algorithm. The proposed crossover and
mutation operators aims to adapt the networks architectures and
weights during the evolution process. Through a specific inheritance
procedure, the weights are transmitted from the parents to their
offsprings, which allows re-exploitation of the already trained
networks and hence the acceleration of the global convergence of the
algorithm. In the preprocessing phase, a new feature extraction
method is proposed based on Legendre moments with the Maximum
entropy principle MEP as a selection criterion. This allows a global
search space reduction in the design of the networks. The proposed
method has been applied and tested on the well known MNIST
database of handwritten digits.
Abstract: This paper reports the study results on neural network
training algorithm of numerical optimization techniques multiface
detection in static images. The training algorithms involved are scale
gradient conjugate backpropagation, conjugate gradient
backpropagation with Polak-Riebre updates, conjugate gradient
backpropagation with Fletcher-Reeves updates, one secant
backpropagation and resilent backpropagation. The final result of
each training algorithms for multiface detection application will also
be discussed and compared.
Abstract: This paper presents and evaluates a new classification
method that aims to improve classifiers performances and speed up
their training process. The proposed approach, called labeled
classification, seeks to improve convergence of the BP (Back
propagation) algorithm through the addition of an extra feature
(labels) to all training examples. To classify every new example, tests
will be carried out each label. The simplicity of implementation is the
main advantage of this approach because no modifications are
required in the training algorithms. Therefore, it can be used with
others techniques of acceleration and stabilization. In this work, two
models of the labeled classification are proposed: the LMLP
(Labeled Multi Layered Perceptron) and the LNFC (Labeled Neuro
Fuzzy Classifier). These models are tested using Iris, wine, texture
and human thigh databases to evaluate their performances.
Abstract: Minimization methods for training feed-forward networks with Backpropagation are compared. Feedforward network training is a special case of functional minimization, where no explicit model of the data is assumed. Therefore due to the high dimensionality of the data, linearization of the training problem through use of orthogonal basis functions is not desirable. The focus is functional minimization on any basis. A number of methods based on local gradient and Hessian matrices are discussed. Modifications of many methods of first and second order training methods are considered. Using share rates data, experimentally it is proved that Conjugate gradient and Quasi Newton?s methods outperformed the Gradient Descent methods. In case of the Levenberg-Marquardt algorithm is of special interest in financial forecasting.
Abstract: Security has been an important issue and concern in the
smart home systems. Smart home networks consist of a wide range of
wired or wireless devices, there is possibility that illegal access to
some restricted data or devices may happen. Password-based
authentication is widely used to identify authorize users, because this
method is cheap, easy and quite accurate. In this paper, a neural
network is trained to store the passwords instead of using verification
table. This method is useful in solving security problems that
happened in some authentication system. The conventional way to
train the network using Backpropagation (BPN) requires a long
training time. Hence, a faster training algorithm, Resilient
Backpropagation (RPROP) is embedded to the MLPs Neural
Network to accelerate the training process. For the Data Part, 200
sets of UserID and Passwords were created and encoded into binary
as the input. The simulation had been carried out to evaluate the
performance for different number of hidden neurons and combination
of transfer functions. Mean Square Error (MSE), training time and
number of epochs are used to determine the network performance.
From the results obtained, using Tansig and Purelin in hidden and
output layer and 250 hidden neurons gave the better performance. As
a result, a password-based user authentication system for smart home
by using neural network had been developed successfully.
Abstract: Although backpropagation ANNs generally predict
better than decision trees do for pattern classification problems, they
are often regarded as black boxes, i.e., their predictions cannot be
explained as those of decision trees. In many applications, it is
desirable to extract knowledge from trained ANNs for the users to
gain a better understanding of how the networks solve the problems.
A new rule extraction algorithm, called rule extraction from artificial
neural networks (REANN) is proposed and implemented to extract
symbolic rules from ANNs. A standard three-layer feedforward ANN
is the basis of the algorithm. A four-phase training algorithm is
proposed for backpropagation learning. Explicitness of the extracted
rules is supported by comparing them to the symbolic rules generated
by other methods. Extracted rules are comparable with other methods
in terms of number of rules, average number of conditions for a rule,
and predictive accuracy. Extensive experimental studies on several
benchmarks classification problems, such as breast cancer, iris,
diabetes, and season classification problems, demonstrate the
effectiveness of the proposed approach with good generalization
ability.
Abstract: Motor imagery classification provides an important basis for designing Brain Machine Interfaces [BMI]. A BMI captures and decodes brain EEG signals and transforms human thought into actions. The ability of an individual to control his EEG through imaginary mental tasks enables him to control devices through the BMI. This paper presents a method to design a four state BMI using EEG signals recorded from the C3 and C4 locations. Principle features extracted through principle component analysis of the segmented EEG are analyzed using two novel classification algorithms using Elman recurrent neural network and functional link neural network. Performance of both classifiers is evaluated using a particle swarm optimization training algorithm; results are also compared with the conventional back propagation training algorithm. EEG motor imagery recorded from two subjects is used in the offline analysis. From overall classification performance it is observed that the BP algorithm has higher average classification of 93.5%, while the PSO algorithm has better training time and maximum classification. The proposed methods promises to provide a useful alternative general procedure for motor imagery classification
Abstract: The quick training algorithms and accurate solution
procedure for incremental learning aim at improving the efficiency of
training of SVR, whereas there are some disadvantages for them, i.e.
the nonconvergence of the formers for changeable training set and
the inefficiency of the latter for a massive dataset. In order to handle
the problems, a new training algorithm for a changeable training
set, named Approximation Incremental Training Algorithm (AITA),
was proposed. This paper explored the reason of nonconvergence
theoretically and discussed the realization of AITA, and finally
demonstrated the benefits of AITA both on precision and efficiency.
Abstract: In this paper, Differential Evolution (DE) algorithm, a new promising evolutionary algorithm, is proposed to train Radial Basis Function (RBF) network related to automatic configuration of network architecture. Classification tasks on data sets: Iris, Wine, New-thyroid, and Glass are conducted to measure the performance of neural networks. Compared with a standard RBF training algorithm in Matlab neural network toolbox, DE achieves more rational architecture for RBF networks. The resulting networks hence obtain strong generalization abilities.
Abstract: In this study a neural network (NN) was proposed to
predict the sorption of binary mixture of copper-cobalt ions into
clinoptilolite as ion-exchanger. The configuration of the
backpropagation neural network giving the smallest mean square
error was three-layer NN with tangent sigmoid transfer function at
hidden layer with 10 neurons, linear transfer function at output layer
and Levenberg-Marquardt backpropagation training algorithm.
Experiments have been carried out in the batch reactor to obtain
equilibrium data of the individual sorption and the mixture of coppercobalt
ions. The obtained modeling results have shown that the used
of neural network has better adjusted the equilibrium data of the
binary system when compared with the conventional sorption
isotherm models.
Abstract: Drying characteristics of rough rice (variety of lenjan) with an initial moisture content of 25% dry basis (db) was studied in a hot air dryer assisted by infrared heating. Three arrival air temperatures (30, 40 and 500C) and four infrared radiation intensities (0, 0.2 , 0.4 and 0.6 W/cm2) and three arrival air speeds (0.1, 0.15 and 0.2 m.s-1) were studied. Bending strength of brown rice kernel, percentage of cracked kernels and time of drying were measured and evaluated. The results showed that increasing the drying arrival air temperature and radiation intensity of infrared resulted decrease in drying time. High bending strength and low percentage of cracked kernel was obtained when paddy was dried by hot air assisted infrared dryer. Between this factors and their interactive effect were a significant difference (p
Abstract: Gasoline Octane Number is the standard measure of
the anti-knock properties of a motor in platforming processes, that is
one of the important unit operations for oil refineries and can be
determined with online measurement or use CFR (Cooperative Fuel
Research) engines. Online measurements of the Octane number can
be done using direct octane number analyzers, that it is too
expensive, so we have to find feasible analyzer, like ANFIS
estimators.
ANFIS is the systems that neural network incorporated in fuzzy
systems, using data automatically by learning algorithms of NNs.
ANFIS constructs an input-output mapping based both on human
knowledge and on generated input-output data pairs.
In this research, 31 industrial data sets are used (21 data for training
and the rest of the data used for generalization). Results show that,
according to this simulation, hybrid method training algorithm in
ANFIS has good agreements between industrial data and simulated
results.