CHAPTER 16

HYBRID SOFT
COMPUTING
TECHNIQUES

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
HYBRID SYSTEMS

Neural Network Systems

Neural networks are the simplified models of the human
nervous systems mimicking our ability to adapt to certain
situations and to learn from the past experiences.

Fuzzy Logic
Fuzzy logic or fuzzy systems deal with uncertainty or
vagueness existing in a system and formulating fuzzy rules
to find a solution to problems.

Genetic Algorithm
Genetic algorithms inspired by the natural evolution process
are adaptive search and optimization algorithms.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
The main aim of the concept of hybridization is to overcome
the weakness in one technique while applying it and bringing
out the strength of the other technique to find solution by
combining them.

Neural networks are good at recognizing patterns but they are

not good at explaining how they reach their decisions.

On the contrary, fuzzy logic is good at explaining the

decisions but cannot automatically acquire the rules used for
making the decisions. Also, the tuning of membership
functions becomes an important issue in fuzzy modeling.
Genetic algorithms offer a possibility to solve this problem.

These limitations act as a central driving force for the creation

of hybrid soft computing systems where two or more
techniques are combined in a suitable manner that
overcomes the limitations of individual techniques.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
The use of hybrid systems is growing rapidly with successful
applications in areas such as

engineering design
stock market analysis and prediction
medical diagnosis
process control
credit card analysis and
few other cognitive simulations.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
 Hybridization Architecture

• Sequential Hybrid Systems

• Auxiliary Hybrid Systems
• Embedded operators

Problem Problem Problem

Algorithm A Algorithm A Algorithm A

Algorithm B Algorithm B Algorithm B

5
Sequential Hybrid System

• Pipeline
• Output of One Sub-system becomes Input to
the other Sub-systems
• Weakest form of Hybridization
• Example: Use one technique to find
promising solutions to a problem and use
another technique to improve the solution
further, like ANN can be used to generate
promising initial candidate solution for GA
April 2007 6
Auxiliary Hybrid System

• One technology calls another technology as a

subroutine
• Example: Structure and Weight Training of
ANN through GA

April 2007 7
Embedded Hybrid System

• Technologies are intertwined.

• Fusion is so complete that it is hard to
segregate them.
• Example: A NN-FL system that receives fuzzy
inputs and processes them to give fuzzy
ouput.

April 2007 8
VARIOUS HYBRID SYSTEMS
In this text book, the following three different hybrid systems are
discussed:

 Neuro fuzzy hybrid system;

 neuron genetic hybrid system;

 fuzzy genetic hybrid systems.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
NEURO FUZZY HYBRID SYSTEMS
Definition:

A neuro-fuzzy hybrid system (also called fuzzy neural hybrid) is a

learning mechanism that utilizes the training and learning algorithms
from neural networks to find parameters of a fuzzy system.

Advantages of neuro fuzzy hybrid systems:

It can handle any kind of information (numeric, linguistic, logical,

etc.).
It can manage imprecise, partial, vague or imperfect information.
It can resolve conflicts by collaboration and aggregation.
It has self-learning, self-organizing and self-tuning capabilities.
It doesn’t need prior knowledge of relationships of data.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
ARCHITECTURE OF NEURO FUZZY HYBRID
SYSTEMS
The general architecture of neuro-
fuzzy hybrid system is as shown in
Figure 1.

Fig 1

The architecture is a three-layer feed forward neural network model.

It can also be observed that the first layer corresponds to the input
variables, and the second and third layers correspond to the fuzzy
rules and output variables, respectively. The fuzzy sets are
converted to (fuzzy) connection weights.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
TYPES OF NEURO FUZZY HYBRID SYSTEMS
NFSs can be classified into the following two systems:

1.Cooperative NFSs.

2. General neuro-fuzzy hybrid systems.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
CO-OPERATIVE NEURO FUZZY SYSTEMS
In this type of system, both artificial neural network (ANN) and fuzzy
system work independently from each other. The ANN attempts to
learn the parameters from the fuzzy system.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
GENERAL NEURO FUZZY HYBRID SYSTEMS
General neuro-fuzzy hybrid systems (NFHS) resemble neural
networks where a fuzzy system is interpreted as a neural network of
special kind.
Figure 2 illustrates an NFHS

In Fig 2, the rule base of

a fuzzy system is
assumed to be a neural
network; the fuzzy sets
are regarded as weights
and the rules and the
input and output
variables as neurons.

Fig 2
“Principles of Soft Computing, 2nd Edition”
by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
GENERAL NEURO HYBRID SYSTEMS
Definition:
A neuro-genetic hybrid or a genetic_neuro-hybrid system is one in
which a neural network employs a genetic algorithm to optimize its
structural parameters that define its architecture.

Properties of genetic neuro-hybrid systems:

1.The parameters of neural networks are encoded by genetic
algorithms as a string of properties of the network, that is,
chromosomes. A large population of chromosomes is generated,
which represent the many possible parameter sets for the given
neural network.

2. Genetic Algorithm _Neural Network, or GANN, has the ability to

locate the neighborhood of the optimal solution quickly, compared to
other conventional search strategies.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
BLOCK DIAGRAM OF GENETIC NEURO
HYBRID SYSTEMS

Fig 3

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
ADVANTAGES OF GENETIC NEURO HYBRID
SYSTEMS

The various advantages of neuro-genetic hybrid are as follows:

GA performs optimization of neural network parameters with

simplicity, ease of operation, minimal requirements and global
perspective.

GA helps to find out complex structure of ANN for given input and
the output data set by using its learning rule as a fitness function.

Hybrid approach ensembles a powerful model that could

significantly improve the predictability of the system under
construction.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
GENETIC FUZZY HYBRID SYSTEMS
The hybridization of genetic algorithm and fuzzy logic can be
performed in the following two ways:

1. By the use of fuzzy logic based techniques for improving genetic

algorithm behavior and modeling GA components. This is called fuzzy
genetic algorithms (FGAs).

2. By the application of genetic algorithms in various optimization and

search problems involving fuzzy systems.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
ADVANTAGES OF GENETIC FUZZY HYBRID
SYSTEMS

GAs allow us to represent different kinds of structures, such as

weights, features together with rule parameters, etc., allowing us to
code multiple models of knowledge representation. This provides a
wide variety of approaches where it is necessary to design specific
genetic components for evolving a specific representation.

Genetic algorithm efficiently optimizes the rules, membership

functions, DB and KB of fuzzy systems. The methodology adopted is
simple and the fittest individual is identified during the process.

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.
 SUMMARY
This chapter has given an overview on Hybrid Soft Computing
Techniques

“Principles of Soft Computing, 2nd Edition”

by S.N. Sivanandam & SN Deepa
Copyright  2011 Wiley India Pvt. Ltd. All rights reserved.