Fuzzy Genetic

Algorithm

Presented By: Dr. Richa

1
 Introduction

 Fuzzy logic

 Genetic Algorithm

 Fuzzy Genetic Algorithm

 Different FGA Approach

 Application Sector

2
 After scientists became disillusioned with classical and neo-
classical attempts at modeling intelligence, they looked in
other directions.
 Two prominent fields arose, connectionism (neural networking,
parallel processing) and evolutionary computing.
 It is the latter that this essay deals with - genetic algorithms
and genetic programming.
 Fuzzy logic is a form of many-valued logic
 A Fuzzy Genetic Algorithm (FGA) is considered as a GA that
uses fuzzy logic based techniques

3
 Definition of fuzzy
 Fuzzy – “not clear, distinct, or precise; blurred”
 Definition of fuzzy logic
 A form of knowledge representation suitable for
notions that cannot be defined precisely, but
which depend upon their contexts.
 Compared to traditional binary sets fuzzy logic
variables may have a truth value that ranges in
degree between 0 and 1
Membership Function
 The membership function represents the
degree of truth as an extension of
valuation.
4
 The term "fuzzy logic" was introduced with
the 1965 proposal of fuzzy set theory by
Lotfi A. Zadeh.
 Fuzzy logic has been applied to many fields,
from control theory to artificial intelligence.
 Fuzzy logics however had been studied
since the 1920s as infinite-valued logics
notably by Łukasiewicz and Tarski.

5
 A point on that scale has three "truth values"—one for each of the
three functions.
 red arrow points to zero, this temperature may be interpreted as
"not hot“
 The orange arrow (pointing at 0.2) may describe it as "slightly
warm“
 The blue arrow (pointing at 0.8) "fairly cold"

6
 A genetic algorithm (or GA) is a search technique used in
computing to find true or approximate solutions to
optimization and search problems.
 Genetic algorithms are categorized as global search heuristics.
 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).

7
 The new population is then used in the next iteration of the
algorithm.
 Commonly, the algorithm terminates when either a maximum
number of generations has been produced, or a satisfactory
fitness level has been reached for the population.
 If the algorithm has terminated due to a maximum number of
generations, a satisfactory solution may or may not have been
reached.
8
Initial Population • The evolution usually starts from a
population of randomly generated
Selection
individuals
Mating • Individual solutions are selected through
a fitness-based process
Crossover
• This generational process is repeated
until a termination condition has been
Mutation
reached.
• improve the solution through repetitive
Terminate application of the mutation, crossover,
inversion and selection operators
9
 The use of FL based techniques for either improving GA behaviour and
modeling GA components, the results obtained have been called fuzzy
genetic algorithms (FGAs),

 The application of GAs in various optimization and search problems

 An FGA may be defined as an ordering sequence of instructions in which

 A fuzzy fitness finding mechanism guides the GA through the search

10
A single objective optimization model cannot serve the purpose of a fitness
measuring index because we are looking at multiple criteria that could be
responsible for stringing together data items into clusters. This is true; not
only for the clustering problem but for any problem solving using GA that
involves multiple criteria. In multi-criteria optimization, the notion of
optimality is not clearly defined. A solution may be best w.r.t. one criterion
but not so w.r.t. the other criteria. Pareto optimality offers a set of non-
dominated solutions called the P-optimal set where the integrity of each of
the criteria is respected.

11
The algorithm has two computational elements that work together.
i) The Genetic Algorithm (GA) and
ii) The Fuzzy Fitness Finder (FFF).

12
 Cossover is a genetic operator used
to vary the programming of a
chromosome or chromosomes from
one generation to the next. It is
analogous to reproduction and
biological crossover, upon which
genetic algorithms are based. Cross
over is a process of taking more than
one parent solutions and producing a
child solution from them.

13
 Mutation is a genetic operator used to maintain genetic diversity
from one generation of a population of genetic algorithm
chromosomes to the next.

 It is analogous to biological mutation. Mutation alters one or

 In mutation, the solution may change entirely from the previous

 Mutation occurs during evolution according to a user-definable

 This probability should be set low. If it is set too high, the search
will turn into a primitive random search.

14
15
 FGA

Fuzzy GA
· A genetic representation for
potential solutions to the problem.
While the population of the genetic · Method to create an initial
algorithm undergoes evolution at population of potential solutions
every generation, the relatively
· An evaluation function to rate
‘good’ solutions reproduce while the
solutions in terms of their “fitness”
relatively ‘bad’ solutions die.
To distinguish between solutions, an · Selection of individuals for the next
generation
objective (evaluation) function is
used. In the simple cases, there is · Genetic operators that alter the
only one criterion for optimization composition of the children
for example, maximization of profit In order to make a successful run of a
or minimization of cost. GA, the values for the parameters of
But in many real-world decision the GA have to be defined like the
making problems, there is a need for population size, parameters for the
simultaneous optimization of genetic operators and the terminating
multiple objectives. condition.
16
• The Fuzzy Fitness Finder
• Input and Output Criteria

• Fuzzification of Inputs
• Fuzzy Inference Engine

• Defuzzification of Output

17
Pittsburgh Approach

Iterative Rule Learning Approach

Michigan Approach

The Nagoya Approach

18
Rough set Approach
 A rough set is a formal approximation of
a crisp set in terms of a pair of sets which
give the lower and the upper
approximation of the original set.
 The tuple composed of the lower and
upper approximation is called a rough
set.
Applications
One of the new data mining theories is the rough set
theories that can be used for
1.Classification to discover structured relationship
within noisy data.
2.Attributes subset selection.
3.Reduction of data set.
4.Finding hidden data patterns
5. Generation of decision rules
 Electrical Engg.

Mechanical Engg.

Economics

Artificial Intelligence

Approx. in all sectors of life.

21
22