Optimized Parameter of Wavelet Neural Network (WNN) using INGA

International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 8 128 – 131
_______________________________________________________________________________________________
128
IJRITCC | August 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
Optimized Parameter of Wavelet Neural Network (WNN) using INGA
Nithya.S[1]
Assistant Professor,
Department Of Computer Science,
Sri Akilandeswari Women’s College,
Vandavasi
Subhashini.R[2]
MSC Computer Science,
Sri Akilandeswari Womens College,
Vandavasi
Sudha.D[3]
MSC Computer Science,
Sri Akilandeswari Women’s College
Vandavasi
Abstract: Genetic algorithm has been one of the most popular methods for many challenging optimization problems. It is a critical problem in
which the evacuation time is an important issues. The continuous air traffic growth and limits of resources, there is a need for reducing the
congestion of the airspace system. The main objective of this work is to automatically adapt the airspace configurations, according to the
evolution of traffic Niche genetic algorithm(INGA) was used in reliability optimization of software system. And also the searching performance
of the genetic algorithm was improved by the stochastic tournament model. The multi-module complex software system reliability allocation
effectively. Genetic algorithm (GA) and FGA are compared though seven benchmark function. It can be applied to a wider range of problem
including multi-level problem. The uniform schema crossover operator and the non-uniform mutation in the genetic algorithm.
Keywords: INGA,GA,WNN,FGA,etc
__________________________________________________*****_________________________________________________
I. INTRODUCTION
1. NEURAL NETWORKS:
 A neural network can be defined as a model of
reasoning based on the human brain. The brain
consists of a densely interconnected set of nerve
cells, or basic information-processing units, called
neurons.[1][4]
 The human brain incorporates nearly 10 billion
neurons and 60 trillion connections, synapses,
between them. By using multiple neurons
simultaneously, the brain can perform its functions
much faster than the fastest computers in existence
today.
 Each neuron has a very simple structure, but an
army of such elements constitutes a tremendous
processing power.
 A neuron consists of a cell body, soma, a number
of fibers called dendrites, and a single long fiber
called the axon.
 Our brain can be considered as a highly complex,
non-linear and parallel information-processing
system.
2. PRECEPTONS:
The operation of Rosenblatt’s perceptron is based
on the McCulloch and Pitts neuron model. The model
consists of a linear combiner followed by a hard limiter.
The weighted sum of the inputs is applied to the
hard limiter, which produces an output equal to +1 if its
input is positive and -1 if it is negative.
In the case of an elementary perceptron, the n-
dimensional space is divided by a hyperplane into two
decision regions. The hyperplane is defined by the linearly
separable function
3. GENETIC ALGORITHM:
A genetic algorithm (or GA) is a search technique
used in computing to find true or approximate solutions to
optimization and search problems[2].
I. Genetic algorithms are categorized as global search
heuristics.
II. Genetic algorithms are a particular class of evolutionary
algorithms that use techniques inspired by evolutionary
biology such as inheritance, mutation, selection, and
crossover (also called recombination).
III.Typically, there will be more than one output class or
class value and therefore more than one output node.
4 . WAVELET NEURAL NETWORK:
A wavelet neural network (WNN) based on the
genetic algorithm (GA) was used in conjunction with an
experimental design to optimize the enzymatic reaction
conditions for the preparation of esteramines-based
esterquats.[3] A set of experiments was designed by central
composite design to process modeling and statistically
evaluate the findings. Five independent process variables,
0
1


n
i
ii wx

Volume: 5 Issue: 8 128 – 131
_______________________________________________________________________________________________
129
_______________________________________________________________________________________
including enzyme amount, reaction time, reaction
temperature, substrates molar ratio and agitation speed were
studied under the given conditions designed by Design
Expert software. All these show that the WNN has great
potential ability in prediction of reaction conversion in
lipase-catalyzed synthesis of products.
Fig1.1 State diagram
5 .OPTIMIZATION PARAMETER:
A optimization is a process that find a best , or
optimal , solution for the problem . the optimization have
major three factor
1. An objective function
2. A set of unkown or variable
3. A set of constraint
5.1 OPTIMIZATION METHODS:
There are two optimization methods are having the
model[1][4]. They are
1. Linear programming.
2. Non linear programming.
Fig 2.1 optimistic model
III. LITERATURE REVIEW
1. INGA-WNN prediction model
1.1 WNN modeling
WNN is generally divided into relax-type and close-type
by the different wavelet-basis functions. And our prediction
adopts the close-type with a three-layer feed-forward neural
network. In our model, we assume that the network topology
appear [5][9], and there are m nodes in the input layer, h
nodes in the hidden layer and n nodes in the output layer.
The input samples are denoted by X1~Xm, the output samples
are denoted by Y1~Yn. The stretching and translating
parameters are denoted by a1~ah, b1~bhrespectively. The
link-weights in the network between theinput layer to
hidden layer and the hidden layer to output layer are denoted
by w~w
, w
′~w
′ respectively.
According to Kolmogorov theory, the number of hidden
layer nodes h is determined by the number of input layer
nodes m. The calculation equation is
h =2 m +1
The sigmoid function is still used as the activation function
shown as follows
g (x ) =1/[1+exp( − x)]
where x in (1)and (3) denotes the data of the previous
layer. By the wavelet basis function, the output results
of WNN are
h ∑r
m
=1wrs − bs
Yt=∑ wst′ψ
a
ss=1
Usually, the gradient descent method is used to calculate
the network connection weights, by which the parameters in
WNN are more accurate than before. However, the extreme
value point from this method is just the approximate value.
Meanwhile, it is vul-nerable to the local optimal network
problem[4]. So we prepare to use the Improved Niche
Genetic Algorithm (INGA) for optimizing the local network
parameters.
In fact, it is expected that a teacher should start from the
familiar concepts in classical mechanics and should land
down with derivation of these equations giving the physical
significance meaningfully. Same is the approach followed
in many of the standard books. Many researchers,

Volume: 5 Issue: 8 128 – 131
_______________________________________________________________________________________________
130
_______________________________________________________________________________________
educationalists have attempted to bridge this gap and tried
to give derivations of SE starting from many fundamental
principles. However, again many of such attempts are
thorny for undergraduate students. [1-9].
1.2 Time dependent Schrödinger equation (TDSE)
The concept of time dependent Schrödinger
equation (TDSE) illustrated in literature and even during
class room teaching is mostly either complex or meant for
advanced learners. This article is intended to enlighten the
concept to the beginners in the field and further to improve
knowledge about detailed steps for abstract mathematical
formulation used which helps in understanding to derive
TDSE using various tools and in more comprehensible
manner. It is shown that TDSE may be derived using wave
mechanics, time independent equation, classical &
Hamilton-Jacobi’s equations[8]. Similar attempts have been
done earlier by some researchers. However, this article
provides a comprehensive, lucid and well derived
derivation, derived using various approaches, which would
make this article unique.
Fig1.3 time dependent flowchart
2. INGA algorithm
Based on the principle of natural selection also referred
to as survival of the fittest, the Genetic algorithm (GA)
originated in the 1960s, which is a global optimization
algorithm[6]. GA is a probability optimization algorithm,
and has great advantages in terms of parallelism and
expansibility. So the algorithm is a good option for dealing
with nonlinear problems. We can quickly obtain the global
optimal solution for large-scale networks.
GA is mainly used to simulate the survival process of
chromosomes. The process includes five steps: encoding,
the determination of the fitness function, the genetic
selection, crossover and mutation. Encoding makes the
problem in the form of code and presents it to the computer.
This paper adopts the real-number encoding. After
completing encoding, this algorithm realizes a series of
operators such as selection, crossover and mutation.
Through the comparison of fitness and Iterative
optimization, we will find the code of the optimal
solution[7].
fig 2.1 inga graph states
Fitness function
The increasing fitness value determines the evolutionary
di-rection of GA. The error E of WNN is defined as the
individual fitness function[8][9]:
E
=
1
∑ (Yt−Yt′)2
2
t=1
IV.ADVANTAGE AND DISADVANTAGE:
ADVANTAGE:
 It improved niche technology
 It improves the convergence speed
 To solve the low convergence fault
DISADVANTAGE:
 It is difficult to preprocess the data.
CONCLUSION:
This paper presents a framework that combiner
similar days election, wavelet, decomposition and neural
network to forecasting. The prediction model combine the
improved niche technology with the flexible genetic
algorithm. Hence the niche automation has the improved
development capacity of genetic algorithm and the
concurrence speed. Moreover, the improvement of
diversification of population solves the problems of the
immature and lower concurrence absolutely .The
Experimentation verified the accuracy and success of the

Volume: 5 Issue: 8 128 – 131
_______________________________________________________________________________________________
131
_______________________________________________________________________________________
INGA which also more absolutely in conclude the network
security satiation .
REFERENCES:
[1] Varun Dutt Indian Institute of Technology, Mandi
Young-Suk Ahn Carnegie Mellon University Cleotilde
Gonzalez Carnegie Mellon University,”Cyber situation
awareness:modelling-research showcase@cmu”
https://repository.cmu.edu//viewcontent.
[2] X.J. Ding, Y. Tian, Y. Yu, A real-time big data gathering
algorithm based on indoor wireless sensor networks for
risk analysis of industrial operations, IEEETrans. Ind.
Inf. 12 (3) (2016) 1232–1242.
[3] Samir W.Mahfound . Report no:95001 ,“Niche methods
for genetic algorithm”
https://www.sciencedirect.com//article//pdf.
[4] Nilesh P. BARDE1, Sandeep D. PATIL,”Fast algorithm
for detecting community structure in networks”
.https://www.ieee.org.
[5] “Exploting time series analysis in twitter to measure a
campaign process”
https://www.ieeeconfpublishing.com
[6] “A network security situation prediction model based on
wavelet neural
network”https://www.sciencedirect.com//article//pdf.
[7] Mark Newman, “Dynamic programming algorithm
optimization for spoken word
recognition”https://www.ieeexplore.ieee.org//document/
11.pdf.
[8] Hamid reza fard masoumi, ”Optimization of process
parameter for lipase-catalyzed synthesis of esteramines-
based esterquats using wavelet neural network”.
https://www.sciencedirect.com//article//pdf.
[9] https://www.cs.wmich.edu.

Optimized Parameter of Wavelet Neural Network (WNN) using INGA

More Related Content

What's hot (17)

Similar to Optimized Parameter of Wavelet Neural Network (WNN) using INGA (20)

More from rahulmonikasharma (20)

Recently uploaded (20)

Optimized Parameter of Wavelet Neural Network (WNN) using INGA