Course Code : CCS364

Course Name : SOFT COMPUTING

Year / Semester : III / V
Department : CSE
Faculty Name : Ms. D.A. Agnes Archana
Regulations : AU-R2021
Academic year : 2023-2024 ODD
 UNIT-I
INTRODUCTION TO SOFT COMPUTING AND FUZZY LOGIC
Introduction - Fuzzy Logic - Fuzzy Sets, Fuzzy Membership Functions, Operations on Fuzzy Sets,
Fuzzy Relations, Operations on Fuzzy Relations, Fuzzy Rules and Fuzzy Reasoning, Fuzzy
Inference Systems

1. SOFT COMPUTING
The term soft computing was coined by Zadah as a multidisciplinary system that combines
the fuzzy logic, neural networks, genetic algorithms etc.

“Soft Computing is an emerging approach to computing which paral els the

remarkable ability of the human mind to reason and learn in an environment of
uncertainty and imprecision”.
The prominent feature of soft computing is its tolerance to imprecise data, uncertainty,
approximate reasoning and partial truth. Soft computing methods mimic the human ability to take
decisions to real world problems. Some of the common soft computing techniques are:
• Fuzzy computing
• Evolutionary computing
• Neural computing
• Probabilistic computing

Soft Computing is the fusion of methodologies designed to model and enable solutions to real
world problems, which are not modelled or too difficult to model mathematically. The aim of Soft
Computing is to exploit the tolerance for imprecision, uncertainty, approximate reasoning, and
partial truth in order to achieve close resemblance with human like decision making.

SOFT COMPUTING CONSTITUENTS

The Soft Computing consists of several computing paradigms mainly:
Neural Networks : for learning and adaptation
Fuzzy set : for knowledge representation via fuzzy If – Then rules
Genetic Algorithms and simulated annealing: for evolutionary computation
These methodologies form the core of SC. Hybridization of these three creates a successful
synergic effect; that is, hybridization creates a situation where different entities cooperate
advantageously for a final outcome. Soft Computing is still growing and developing. Hence, a
clear definite agreement on what comprises Soft Computing has not yet been reached. More new
sciences are still merging into Soft Computing.

FROM CONVENTIONAL AI TO COMPUTATIONAL INTELLIGENCE:

Humans usually employ natural languages in reasoning and drawing Conclusions.
Conventional AI focuses on an attempt to mimic human intelligent behaviour by expressing it in
language forms or symbolic forms.
 Fig 1.1 An expert system: one of the successful (conventional) AI products

It basically manipulates symbols on the assumption that such behaviour can be stored in
symbolically structured knowledge base. This is so-called physical symbol system hypothesis. The
most successful conventional AI product is the knowledge-bases system or expert system (ES)
represented in Fig 1.1
AI field is steadily expanding; the boundary between AI and soft computing is becoming
indistinct and obviously, successive generations of AI methodologies will be growing more
sophisticated. The long-term goal of AI research is the creation and understanding of machine
intelligence. An intelligence system represented in Fig 1.2 can sense its environment (perceive)
and act on its perception (react).

Fig 1.2 Intelligent system

An easy extension of ES may also result in the same ideal computationally intelligent system
sought by soft computing researchers. Soft computing is apparently evolving under AI influences
that sprang from cybernetics (the study of information and control in humans and machines).
1.1 INTRODUCTION
Fuzzy logic is a superset of conventional (or Boolean) logic and contains similarities and
differences with Boolean logic. Fuzzy Logic is similar to Boolean logic in that Boolean logic
results are returned by FL operations when all fuzzy memberships are restricted to 0 and 1. Fuzzy
Logic differs from Boolean logic in that it is permissive of natural language queries and is more
like human thinking; it is based on degrees of truth. For example, traditional sets include or do not
include an individual element; there is no other case than true or false. However, fuzzy set allow
partial membership.
FL is basically a multi valued logic that allows intermediate values to be defined between
conventional evaluations such as yes/no, true/false, black/white, etc. Notions like rather warm or
pretty cold can be formulated mathematically -and processed with the computer.
In this way, an attempt is made to apply a more human-like way of thinking in the programming
of computers. Fuzzy logic is a problem-solving control system methodology that lends itself to
implementation in systems ranging from simple, small, embedded microcontrollers to large,
networked, multichannel PC or workstation based data acquisition and control systems. It can be
implemented in hardware, software or a combination of both.
FL Provides a simple way to arrive at a definite conclusion based upon vague, ambiguous,
imprecise, noisy, or missing input information. FLs approach to control problems mimics how a
person would make decisions, only much faster.

Figure. 1.3 Fuzzy Logic System

A fuzzy logic system accepting imprecise data and providing a decision. Fuzzy sets that
represent fuzzy logic provide means to model the uncertainty associated with vagueness,
imprecision and lack of information regarding a problem or a plant or a system, etc.

Classical Sets (Crisp Sets)

Basically, a set is defined as a collection of objects, which share certain characteristics. A
classical set is a collection of distinct objects. For example, the user may define a classical set of
negative integers, a set of persons with height less than 6 feet, and a set of students with passing
grades. Each individual entity in a set is called a member or an element of the set. The classical set
is defined in such a way that the universe of disclosure is splitted into two group: members and
non-members. Consider an object x in a crisp set A. This object x is either a member or a on
member of the given set A. In case of crisp sets, no partial membership exists. A crisp set is defined
by its characteristic function. Let universe of discourse be U. The collection of elements in the
universe is called whole set. The total number of elements in universe U is called cardinal number
denoted by nU.
 Collections of elements within a universe are called sets, and collections of elements
within a set are called subsets.
  We know that for a crisp set A in universe U:
1. An object x is a member of given set A (x ε A), i.e., x belongs to A.
2. An object x is not a member of given set A (x Ɇ A), i.e., x does not belong to A.
Fuzzy Sets
Fuzzy sets may be viewed as an extension and generalization of the basic concepts of crisp
sets. An important property of fuzzy set is that it allows partial membership. A fuzzy set is a set
having degrees of membership between 1 and 0.The membership in a fuzzy set need not be
complete, i.e., member of one fuzzy set can a1so be member of other fuzzy sets in the same
universe. Fuzzy sets can be analogous to the thinking of intelligent people. If a person has to be
classified as friend or enemy, intelligent people will not resort to absolute classification as friend
or enemy. Rather, they will classify the person somewhere between two extremes of friendship
and enmity. Similarly, vagueness is introduced in fuzzy set by eliminating the sharp boundaries
that divide members from non-members in the group.
A fuzzy set Ḛ in the universe of discourse U can be defined as a set of ordered pairs and it
is given by
Ḛ = {(𝑥, µḚ(𝑥))|𝑥𝜀𝑈)
There are other ways of representation of fuzzy sets; all representations allow partial
membership to be expressed. When the universe of discourse U is discrete and finite fuzzy set Ḛ is
given as follows:

A fuzzy set is universal fuzzy set if and only if the value of the membership function is 1
for all the members under consideration.

INTRODUCTION TO FUZZY LOGIC (FL)

The concept of Fuzzy Logic (FL) was introduced by LotfiZadeh. Fuzzy logic is seen as
a method of processing data. It is a problem solving strategy involving hardware, software or
even their combination. The fuzzy logic is based on experience and learning; it is not rule based.
It is a multi- valued logic.

Fuzzy Logic recognizes more than true and false values. The propositions in fuzzy
logic can be represented with degrees of truthfulness and falsehood rather than
just as true or false.

Example 1: The rule based and other predictive approaches to the question “Will it rain to-
morrow?” will give the answer either as yes or no. The fuzzy logic wi ll give the answer in more
detailed way as the degree or chance that it will rain is 60% or the probability that it would rain
tomorrow is 0.6.
1.2 MEMBERSHIP FUNCTIONS

The membership function is a graphical representation of the magnitude

of participation of each input.

• It associates a weight to each input that are processed, define functional dependency
between inputs, and finally determines an output response.
• In other words, a membership function (MF) is a curve that defines how each point in the
input space is mapped to a membership value (or degree of membership) between 0 and
1.
Example 2: The following figure gives the representation of the color tones black and white in
Boolean and fuzzy logic.
In Boolean logic, the color is represented either as black or white, whereas in Fuzzy logic
the transition of white to black tones (with increasing grey tones) can be represented. As the
white changes to grey then to black the belongingness also changes gradually from set white to
set black.

Fig 1.4 Comparison of Boolean and Fuzzy Logic

• Determining the right input/output membership functions is the first step of the fuzzy
logic control process.
• A fuzzy algorithm categorises the information entering a system and assigns values that
represent the degree of membership in those categories.
• The Universe of Discourse is the range of all possible values for an input to a fuzzy
system.
• The input membership values are applied on the rules to determine their influence on the
fuzzy output sets of the final output.
• Once the functions are inferred, scaled, and combined, they are defuzzified into a crisp
output which drives the system.
• The crisp output or membership function is either 0 or 1.
• The fuzzy set do not have crisp boundary. It shows the transition from belonging to a set
to not belonging to a set.
• This smooth transition is represented by membership function.
Example 3: Figure 1.5 represents a crisp membership function that gives value 1 if the temperature
is greater than or equal to 80F and 0 if it is less than 80F.

Fig 1.5 Crisp membership function

 The crisp logic function cannot distinguish between members of the same set.
 Input membership functions can be triangles, trapezoids, bell curves or any other shape.
The shapes represent the distribution of information within the system.
 𝜇𝐴(𝑋) is a membership function that represents the categorization of an element X to
set
 The discrete fuzzy set A is given as A={ (X )/X ,  (X )/X  (X )/X }.
A 1 1 A 2 2, …, A n n

A fuzzy set A with universe of discourse X has:

A (X): X[0, 1], where
A (X)=1, if X is completely in A
A (X)=0, if X is completely not in A
0< A(X)<1, if X is partially in A.
Example 4:The following figure shows the membership function of tallness. Determine the
membership of 200cm, 170 cm and 150 cm.

Fig 1.6 Membership function of tallness

tall(200)= 1.
tall (170)=0.2
tall (150)=0
The discrete fuzzy value A={0/150, 0.2/160,1/200}
Common Membership Functions (MF)
The commonly used membership functions are explained.
a) Triangular function
This function is specified using three parameters

Fig 1.7: Triangular Membership

Function
In the above figure, if the value of x <20 the membership function is 0; if the value of x is
between a and b, then the membership function is (x-a) /(b-a) (i.e.) the positive slope of ab; if
the value of x is between b and c, then the membership function is (x-a) /(c-b) (i.e.) the negative
slope of bc; if the value of x is >c, then the membership function is 0.
b) Trapezoidal Function
This function is specified by four parameters.

Fig 1.8 Trapezoidal function

c). Gaussian Function

This is specified by two parameters (m, σ). Here m denote the center and σ denote the width
of the function. A small σ will generate a thin MF, while a big σ will lead to a flat MF.

Fig 1.9 Gaussian Function

c). Bell shapped MF

This function is specified by three parameters {a, b, c}. The parameter b is usually positive
value and we can adjust c and a to vary the center and width of the function. The parameter b is
 used to control the slopes.

Fig 1.10 Bell shapped MF

d). Sigmoidal MF
This function is specified by two parameters a and c, where a controls the slope of the curve
and c is the center of the function.

Fig 1.11 Sigmoidal MF

Comparison of Fuzziness and probability

The fuzziness and probability seem to be very similar. But they have their inherent
differences. Often the density function of probability is confused with the membership functions
of fuzziness.

Differences between probability and fuzziness

Probability Fuzziness
Probability is an aggregated property. Fuzziness is a property of single event.
It is expressed as density function. It is expressed as membership function.
Probability is concerned with the undecid- Fuzzy logic is concerned with the
ability in the outcome of clearly defined ambiguity or undecidability inherent in the
and randomly occurring events. description of the event itself.
Features of membership functions
The membership function is defined by the following three properties:
1. Core:
The area for which elements have maximum degree of membership to the fuzzy set A (i.e.)
1.
µA(x) = 1
2. Support:
The area where the membership function is greater than zero.
µA(x) >0
3. Boundary:
If the region of universe has a nonzero membership but not full membership, this defines
the boundary of a membership.

Fig 1.12: Properties of membership function

Regions in membership function
• Crossover point: The crossover point of a membership function is the elements in
universe whose membership value is equal to 0.5.
µA(x) =0.5
• Height: The height of the fuzzy set A is the maximum value of the membership
function.
Height = max(µA(x))

CLASSICAL SET (Crisp set)

A classical set is a collection of objects with certain characteristics. For example, the user
may define a classical set of negative integers, a set of persons with height less than 6 feet, and a
set of students with passing grades. Each individual entity in a set is called a member or an element
of the set.
There are several ways for defining a set. A set may be defined using one of the following:
1. The list of all the members of a set may be given.
Example A= {2,4,6,8,10} (Roaster form)
2. The properties of the set elements may be specified.
Example A = {x|x is prime number < 20} (Set builder form)
3. The formula for the definition of a set may be mentioned.
Example A = {xi = xi + 1 5 , i = 1 to 10, where xi = 1}
4. The set may be defined on the basis of the results of a logical operation.
Example A = {x|x is an element belonging to P AND Q}
5. There exists a membership function, which may also be used to define a set. The membership
is denoted by the letter 𝜒 and the membership function for a set A is given by (for all values
of x).
𝜒𝐴 (𝑥) = { 1, 𝑖𝑓 𝑥 ∈ 𝐴
0, 𝑖𝑓 𝑥 ∉ 𝐴
The set with no elements is defined as an empty set or null set. It is denoted by symbol Ø.
The set which consist of all possible subset of a given set A is called power set
𝑃(𝐴) = {𝑥|𝑥 ⊆ 𝐴}
1.3 FUZZY OPERATIONS
Fuzzy operations are very useful in the design of a Fuzzy Logic Controller. It allows the
manipulation of fuzzy variables by different means.
Given à and B are the two fuzzy sets, and X be the universe of discourse with the following
respective member functions:

The operations of Fuzzy set are as follows:

Union Operation:
The union operation of a fuzzy set is defined by:

μA𝖴B(x) = max(μA(x),μB(x))
Example:
Let's suppose A is a set which contains following elements:
A = {( X1, 0.6 ), (X2, 0.2), (X3, 1), (X4, 0.4)}
And, B is a set which contains following elements:
B = {( X1, 0.1), (X2, 0.8), (X3, 0), (X4, 0.9)}
then,
 AUB = {( X1, 0.6), (X2, 0.8), (X3, 1), (X4, 0.9)}
Because, according to this operation
For X1
μA𝖴B(X1) = max (μA(X1), μB(X1))
μA𝖴B(X1) = max (0.6, 0.1)
μA𝖴B(X1) = 0.6

For X2
μA𝖴B(X2) = max (μA(X2), μB(X2))
μA𝖴B(X2) = max (0.2, 0.8) μA𝖴B(X2) = 0.8
For X3
A𝖴B(X3) = max (μA(X3), μB(X3))
μA𝖴B(X3) = max (1, 0)
μA𝖴B(X3) = 1
For X4
A𝖴B(X4) = max (μA(X4), μB(X4))
μA𝖴B(X4) = max (0.4,0.9)
μA𝖴B(X4) = 0.9

Intersection Operation:
The intersection operation of fuzzy set is defined by:
μA∩B(x) = min (μA(x), μB(x))
Example:
Let's suppose A and B are a set which contains following elements:
A = {( X1, 0.3 ), (X2, 0.7), (X3, 0.5), (X4, 0.1)} And,
B = {( X1, 0.8), (X2, 0.2), (X3, 0.4), (X4, 0.9)} then
A∩B = {( X1, 0.3), (X2, 0.2), (X3, 0.4), (X4, 0.1)}
Because, according to this operation
For X1
μA∩B(X1) = min (μA(X1), μB(X1))
μA∩B(X1) = min (0.3, 0.8)
μA∩B(X1) = 0.3
For X2
μA∩B(X2) = min (μA(X2), μB(X2))
μA∩B(X2) = min (0.7, 0.2)
μA∩B(X2) = 0.2
For X3
μA∩B(X3) = min (μA(X3), μB(X3))
μA∩B(X3) = min (0.5, 0.4)
μA∩B(X3) = 0.4
For X4
μA∩B(X4) = min (μA(X4), μB(X4))
μA∩B(X4) = min (0.1, 0.9)
μA∩B(X4) = 0.1

3. Complement Operation:
The complement operation of fuzzy set is defined by:
μĀ(x) = 1-μA(x),
Example:
Let's suppose A and B are two sets which contains following elements:
A = {( X1, 0.3 ), (X2, 0.8), (X3, 0.5), (X4, 0.1)}
then,
Ā= {( X1, 0.7 ), (X2, 0.2), (X3, 0.5), (X4, 0.9)}
Because, according to this operation
For X1
 μĀ(X1) = 1-μA(X1)
μĀ(X1) = 1 - 0.3
μĀ(X1) = 0.7
For X2
μĀ(X2) = 1-μA(X2)
μĀ(X2) = 1 - 0.8
μĀ(X2) = 0.2
For X3
μĀ(X3) = 1-μA(X3)
μĀ(X3) = 1 - 0.5
μĀ(X3) = 0.5
For X4
μĀ(X4) = 1-μA(X4)
μĀ(X4) = 1 - 0.1
μĀ(X4) = 0.9

Products
Algebric product or Vector product ( ∙ 𝑩):
𝜇𝐴∙𝐵 𝑥 = 𝜇𝐴 𝑥 ∙ 𝜇𝐵 𝑥
Scalar product ( × 𝑨):
𝜇𝛼𝐴 𝑥 = 𝛼 × 𝜇𝐴

Sum and Difference

Sum (A + B):
𝜇𝐴+𝐵 𝑥 = 𝜇𝐴 𝑥 + 𝜇𝐵 𝑥 − 𝜇𝐴 𝑥 . 𝜇𝐵 𝑥
 Difference (𝐴 − 𝐵 = 𝐴 ∩ ) :
𝜇𝐴−𝐵 𝑥 = 𝜇𝐴∩𝐵𝐶 𝑥
Disjunctive sum:
𝐴⨁𝐵 = (𝐴𝐶 ∩ 𝐵) 𝖴 (𝐴 ∩ 𝐵𝐶 )
Bounded Sum: | 𝐴 𝑥 ⨁𝐵 𝑥 |= 𝜇|(𝑥)⨁𝐵(𝑥)| = min{1, 𝜇𝐴 𝑥 + 𝜇𝐵 𝑥 }
Bounded Difference: 𝐴 𝑥 ⊖ 𝐵 𝑥 = 𝜇|(𝑥)⊖𝐵(𝑥)| = max{0, 𝜇𝐴 𝑥 + 𝜇𝐵 𝑥 − 1}
Equality and Power
Equality (𝐴 = 𝐵): 𝜇𝐴 𝑥 = 𝜇𝐵 𝑥
Power of a fuzzy set : 𝜇𝐴 𝛼 𝑥 = (𝜇𝐴 𝑥) 𝛼
 If α < 1, then it is called dilation
 If α > 1, then it is called concentration

Cartesian product
Cartesian product (𝐴 × 𝐵): 𝜇𝐴×𝐵 𝑥, 𝑦 = min (𝜇𝐴 𝑥, 𝜇𝐵 𝑦)

Example: (𝑥) = {(𝑥1, 0.2), (𝑥2, 0.3), (𝑥3, 0.5), (𝑥4, 0.6)}
(𝑦) = {(𝑦1, 0.8), (𝑦2, 0.6), (𝑦3, 0.3)}
𝑦1 𝑦2 𝑦3
𝑥1 0.2 0.2 0.2
A × 𝐵 = min (𝑥) , 𝜇𝐵(𝑦) =𝑥2 0.3 0.3 0.3
𝑥3 0.5 0.5 0.3
𝑥4 0.6 0.6 0.3

Properties of fuzzy sets

Commutativity:
𝐴∩ 𝐵 =𝐵 ∩𝐴
𝐴𝖴B=𝐵𝖴A
Associativity:
𝐴 𝖴 (𝐵 𝖴 𝐶) = (𝐴 𝖴 𝐵) 𝖴 𝐶
𝐴 ∩ (𝐵 ∩ 𝐶) = (𝐴 ∩ 𝐵) ∩ 𝐶
Distributivity :
𝐴 𝖴 (𝐵 ∩ 𝐶) = (𝐴 𝖴 𝐵) ∩ (𝐴 𝖴 𝐶)
𝐴 ∩ (𝐵 𝖴 𝐶) = (𝐴 ∩ 𝐵) 𝖴 (𝐴 ∩ 𝐶)
Idempotence:
𝐴𝖴𝐴=𝐴𝐴∩𝐴=∅;
𝐴 𝖴 ∅; = 𝐴 𝐴 ∩ ∅ ; = ∅;
Transitivity:
If 𝐴 ⊆ 𝐵; 𝐵 ⊆ 𝐶 then 𝐴 ⊆ 𝐶
Involution :
(𝐴 𝑐 ) 𝑐 = 𝐴
De Morgan’s law :
(𝐴 ∩ 𝐵) 𝑐 = 𝐴 𝑐 𝖴 𝐵 𝑐
 (𝐴 𝖴 𝐵) 𝑐 = 𝐴 𝑐 ∩ 𝐵 𝑐

`1.4 CRISP RELATION AND FUZZY RELATION:

Relationships between objects are the basic concepts involved in decision making and other
dynamic system applications. The relations are also associated with graph theory, which has a
great impact on designs and data manipulations. Relations represent mappings between sets and
connectives in logic. A classic binary relation represents the presence or absence of a connection
or interaction or association between the elements of two sets. Fuzzy binary relations are a
generalization of crisp binary relations, and they allow various degrees of
relationship(association)between elements. Ln a fuzzy binary relation, the degree of association is
represented by membership grades in the same way as the degree of set membership is represented
in a fuzzy set.
1.4.1 Cartesian Product of Relation
An ordered r-tuple is an ordered sequence of r-elements expressed in the form (a1, a2, a3,
... ,ar). An unordered r-tuple is a collection of r-elements without any restrictions in order. For r =
2, the r-tuple is called an ordered pair. For crisp sets A1, A2, ... , Ar, the set of all r-tuples (al,
a2,a3, ... , ar), where a1 εA1, a2εA2... , arε Ar, is ca lled the Cartesian product of AI,A2... ,Ar and,
is denoted by. AIX A2X· ... X Ar. The Cartesian product of two or more sets is not the same as
the arithmetic, product of two or more sets. If all the a r's are identical and equal to A, then the
Cartesian product A 1 x A2X · · · x Ar is denoted as Ar.

1.4.2 Classical Relation

An r·ary relation over At.A2, ... ,Ar, is a subset of the Cartesian product A1X A2x · · · x Ar,.
When r = 2, the relation is a subset of the Cartesian product A1 x A2. This is called a binary relation
from A1toA2. When three, four or five sets are involved in the subset of full Cartesian product then
the relations are called ternary, quaternary and quinary, respectively. Generally, the discussions
are centred on binary relations.
Consider two universes X and Y; their Cartesian product X x Y is given by

Here the Cartesian product forms an ordered pair of every xεX with every y εY. Every
element in X is completely related to every element in Y. The characteristic function, denoted by
x, gives the strength of the relationship between ordered pair of elements in each universe. If it
takes unity as its value, then complete relationship is found, if the value is zero, then there is no
relationship i.e.

When the universes or sets are finite, then the relation is represented by a matrix called relation
matrix. An r-dimensional relation matrix represents an r-ary relation. Thus, binary relations are
represented by two-dimensional matrices.
Consider the elements defined in the universes X and Y as follows:
 The Cartesian product of these two sets leads to

From this set one may select a subset such that

Subset R can be represented using a coordinate diagram as shown in Fi re 8-1.

The relation could equivalently be represented using a matrix as allows:

The relation between sets X and Y may also be expressed by mapping representations as
shown in Figure 1.13.

Fig 1.13 Mapping representation of a relation

A binary relation in which each element from first set X is not mapped to more than one element
in second set Y is called a function and is expressed as

Cardinality of Classical Relation

Consider n elements of universe X being related to m elements of universe Y. When the
cardinality ofX=nXand the cardinality of Y = ny, then the cardinality of relation R between the two
universes is
 The cardinality of the power set P(X xY) describing therelation is given by

Operations on Classical Relations

Let Rand S be two separate relations on the Cartesian universe X xY. The null relation and
the complete relation are defined by the relation matrices ɸ R and ER. An example of a 3 X 3 form
of the ɸR and ER matrices is given below:

Function-theoretic operations for the two crisp relations (R, 5) are defined as follows:
1. Union

2. Intersection

3. Complement

4. Containment

5. Identity

Properties of Crisp Relations

 The properties of classical set operations such as commutativity, associativity, distributivity
and involution and idempotency also hold good for classical relations.
 Similarly De Morgan’s law and excluded middle laws hold good for crisp relations as they do
for crisp sets.
 The null relation ɸRis analogous to null set ɸ and complete relation ER is analogous to whole
set X.
1.4.3 Fuzzy Relation
 Fuzzy relations relate elements of one universe (say X) to those of another universe (say Y)
through the Cartesian product of the two universes. These can also be referred to as fuzzy sets
defined on universal sets, which are Cartesian products.
 A fuzzy relation is based on the concept that everything is related to some extent or unrelated.
 A fuzzy relation is a fuzzy set defined on the Cartesian product of classical sets {X1, X2, ...Xn}
where tuples (x1, x2…, xn) may have varying degrees of membership µR(x1, x2…, xn) within
the relation. That is,

 A fuzzy relation between two sets of X and Y is called binary fuzzy relation and is denoted
by R(X, Y).
 A binary relation R(X, Y) is referred to as bipartite graph when X ≠ Y. The binary relation on
a single set X is called directed graph or digraph. This relation occurs when X =Y and is
denoted as R(X,X) or R(X2). Let

Fuzzy relation R(X,Y) can be expressed by an nx m matrix as follows:

The matrix representing a fuzzy relation is called fuzzy matrix.

Cardinality of Fuzzy Relations

The cardinality of fuzzy sets on any universe is infinity; hence the cardinality of a fuzzy
relation between two or more universes is also infinity. This is mainly a result of the occurrence
of partial membership in fuzzy sets and fuzzy relations.

Operations on Fuzzy Relations

The basic operations on fuzzy sets also apply on fuzzy relations. Let E and S, be fuzzy
relations on the Cartesian space X xY. The operations that can be performed on these fuzzy
relations are described below:
1. Union

2. Intersection

3. Complement

4. Containment
Properties of Fuzzy Relations
 Like classical relations, the properties of commutativity, associativity, distributivity,
idempotency and identity also hold good for fuzzy relations.
 DeMorgan's laws hold good for fuzzy relations as they do for classical relations.
 The null relation and complete relation are analogous to the null set and the whole set
respectively, in set theoretic form. The excluded middle laws are not satisfied in fuzzy
relations as for fuzzy set.

 This is because a fuzzy relation is also a fuzzy set, and there exist an overlap, between a
relation and its complement. Hence

1.4.4 Tolerance and Equivalence Relations

 Relations possess various useful properties. Relations play a major role in graph theory.
 The three characteristic properties of relations discussed are: reflexivity, symmetry and
transitivity.
 The anonyms of these properties are: reflexivity, asymmetry and non transitivity.
1.A relation is said to be reflexive if every vertex (node) in the graph originates a single
loop as shown in Figure 1.14 a.
2. A relation is said to be symmetric if for every edge pointing from vertex I to vertex j,
there is an edge pointing in the opposite direction, i.e., from vertex j to vertex i where i,j=
1, 2, 3, .... Figure 1.14 b represents asymmetric relation.
3. A relation is said to be transitive if for every pair of edges in the graph – one pointing
from vertex i to vertex j and the other pointing from vertex j to vertex k, there is an edge
pointing vertex I to vertex k.

a.Reflexive b.Symmetric c.Transitive

Fig 1.14 Three vertex node

Figure 1.14c represents a transitive relation. Here an arrow points from node 1 to node 2 and
another arrow extends from node 2 to node 3. There is also an arrow from node 1 to node 3.
Classical Tolerance Relation
 The tolerance relation R1 on universe X is one where the only properties of reflexivity and
symmetry are satisfied.
 The tolerance relation is called proximity relation.
 An equivalence relation can be formed from tolerance relation R1 by (n-1) compositions
with itself, where n is the cardinality of the set that defines R1, here it is X, i.e.

Fuzzy Equivalence Relation

Let be a fuzzy relation on universe X, which maps elements from X to X Relation will
be a fuzzy equivalence relation if all the three properties - reflexive, symmetry and transitivity -
are satisfied. The membership function theoretic forms for these properties are represented as
follows:

1. Reflexivity

If this is not the case for few x ε X, then R(X,X) is said to be irreflexive.

2. Symmetry

If this is not satisfied for few xi,XjεX, then R(X,X) is called asymmetric.

3. Transitivity

Where

Fuzzy Tolerance Relation

 A binary fuzzy relation that possesses the properties of reflexivity and symmetry is called
fuzzy tolerance relation or resemblance relation.
 The equivalence relations are a special case of the tolerance relation. The fuzzy tolerance
relation can be reformed into fuzzy equivalence relation in the same way as a crisp tolerance
relation is reformed into crisp equivalence relation, i.e.,
Where”n" is the cardinality of the set that defines R1

1.5 FUZZY RULE BASE AND APPROXIMATE REASONING

A production system or a rule-based system has three basic components:
• a set of rules [rule base]
• a data base [fact base]
• interpreter for the rules

Fuzzy inference or fuzzy rule based system adapts the process of mapping from
a given input to an output using fuzzy logic.
The process of mapping can be based on membership functions, fuzzy logic operators or if-
then rules. A database which defines the membership functions of the fuzzy sets are used in the
fuzzy rules.

Fig 1.15 Architecture of Fuzzy rule base

The following are the steps in fuzzy reasoning:

• Fuzzification: Compare the input variables with the membership functions to obtain the
membership values of each linguistic label.
• Combine the membership values on the premise part to obtain the membership degree of
each rule.
• Generate the qualified consequents or each rule depending on the firing strength.
• Defuzzification: Aggregate the qualified consequents to produce a crisp output.
Linguistic Variables
Linguistic variable are variables whose values are words or sentences not numerical values.
They are used to quantify and reason about vague or fuzzy terms that appear in natural language.
The fuzzy numbers that represent linguistic concepts, such as very small, small, medium, and so
on, are usually called linguistic variables.
Examples for linguistic variables:
The temperature values can be hot, cold etc. The temperature is the linguistic variable and
the linguistic values are hot, cold.
The states of linguistic variable are expressed in linguistic terms. These terms are de- fined as
fuzzy numbers in terms of a base variable. The values of the base variable are real numbers within
a specific range. The base variable can be physical quantity like temperature of numerical quantity
like age.
Each linguistic variable is fully characterized by a quintuple (v, T, X, g, m) where v: the
name of the variable
T : the set of linguistic terms of v that refer to a base variable whose values range over a
universal set X
g:a syntactic rule (a grammar) for generating linguistic terms. m : a semantic rule that assigns
to each linguistic term t.

Fig 1.16 Linguistic variable

Truth Tables
Truth tables define logic functions of two propositions. Let X and Y be two propositions,
either of which can be true or false. The operations over the propositions are:
• Conjunction (^): X AND Y.
• Disjunction (v): X OR Y.
• Implication or IF X THEN Y.
• ONLY IF Y.
Fuzzy Rules
A fuzzy rule is defined as the conditional statement of the form “If x is A THEN y is B” where
x and y are linguistic variables and A and B are linguistic values determined by fuzzy sets on the
universes of discourse X and Y. The general form of fuzzy rules is:
IF premise or condition (antecedent), THEN conclusion (consequent)
• The decision-making process in fuzzy is based on rules with sentence conjunctives
AND, OR and ALSO.
• Each rule corresponds to a fuzzy relation. All the rules belong to a rule base.
Formation of fuzzy rules:
There are three general strategies to form fuzzy rules:
• Assignment statements: The linguistic variables are given values.
Example: Climate= summer; temperature=cold
• Conditional statements: Some conditions are given for linguistic variables. The rules
are framed with IF…THEN…ELSE words.
Example: IF the room is very cool THEN turn off the fan ELSE speed up the fan.
 • Unconditional statements:The values for linguistic variables are given without stating
any condition.
Example: Goto product; Divide by a.
Decomposition of Fuzzy rules
The basic properties and operations of fuzzy logic could be exploited to decompose the
compound rules into simple canonical rules. These rules are based on natural language
representations and fuzzy logic. By the process of decomposition, the compound rule is rewritten
as simple rules.
Canonical form:
Rule 1: IF condition C1, THEN restriction R1
Rule 2: IF condition C2, THEN restriction R2
………………………………………………..
Rule n: IF condition Cn, THEN restriction Rn.
• The rules are constructed with linguistic connectors like “AND”, “OR”,”ELSE”.
• The restrictions in the rules are the output or consequent of the rules.
• The fuzzy rules can be decomposed in one of the following methods:
• Multiple conjunctive antecedents
• Multiple disjunctive antecedents
Multiple conjunctive antecedents
• This method is used when the compound rule has so many “AND” connectors in
their antecedent part.
• The general form of the compound rule is:
IF x is A1 and A2 and … and Al THEN y is Bs
• A new fuzzy set As can be formed, which is the conjunction of all the conditions
given in the rule.
As={A1 ∩ A2 ∩ …∩Al}
• The membership function is
𝜇(𝑥) = min[µ𝐴1(𝑥), µ𝐴2 (𝑥),… . . µ𝐴𝑖 (𝑥)]
• Now the compound rule is rewritten as
IF x is As then y is Bs
Multiple disjunctive antecedents
• This method is used when the compound rule has so many “OR” connectors in their
antecedent part.
• The general form of the compound rule is:
IF x is A1 or A2 or … or Al THEN y is Bs
• A new fuzzy set As can be formed, which is the conjunction of all the conditions
given in the rule.
As={A1 U A2 U …UAl}
• The membership function is

• Now the compound rule is rewritten as

IF x is As then y is Bs
Aggregation of Fuzzy Rules
There are two extreme cases:
Aggregation of rules is the process of obtaining the overall consequents from
the individual consequents provided by each rule.
Conjunctive system of rules: This is same as multiple conjunctive antecedents where the rules
are connected by AND connector.
Disjunctive system of rules: This is same as multiple disjunctive antecedents where the rules
are connected by OR connector.

FUZZY PROPOSITIONS
Fuzzy propositions use fuzzy predicates, modifies, quantifiers and qualifiers to express
fuzzy logic. The following are the fuzzy propositions:
• Fuzzy Predicates: Predicates are generally expressed in natural language rather than as
crisp values.
Example: Radha is good.
• Fuzzy predicate modifiers: The modifiers act as hedges. They are used in linguistic
variables. Generally adjectives act as modifiers.
Example: The question paper is very tough. Here ‘very’ is the modifier.
• Fuzzy quantifiers: The quantifiers are used as a measure or cardinality. The commonly
used modifiers are many, most, several, few etc.
Example: Only very few people eat healthy food. Here very few is the modifier.
• Fuzzy qualifiers: There are four qualifiers:
i) Fuzzy truth qualification: This asserts the degree of truth. The general form is “a is
in x”. The proposition ‘a’ has the truth value of ‘x’.
Example: “Iniyan is bad” is not true. Here the statement “Iniyan is bad” is the
proposition and “not true” is the truth value.
ii) Fuzzy probability qualification: The probability can be expressed either as numerical
value or as intervals. The words that qualify the probability are likely, unlikely, very
likely etc. The general form is “ X is Y” where X is the proposition and Y is the fuzzy
value.
iii) Fuzzy usuality qualification: The general form is: “usually (X)=usually (X is F) ”.

propositions with high probability value will fall under this category.
iv) Fuzzy possibility qualification: The words used to express the possibility are
impossible, possible, quite impossible etc. The general form is “X is Y” where the
propositions X takes the value the fuzzy value Y.

1.5.2 FUZZY REASONING

This is otherwise known as approximate reasoning. Fuzzy reasoning for real-world problems
involves missing knowledge, inexact knowledge, inconsistent facts or rules, and other sources
of uncertainty. There are four types of approximate reasoning:
• Categorical reasoning
 • Qualitative reasoning
• Syllogistic reasoning
• Dispositional reasoning
Categorical Reasoning
In categorical reasoning, the antecedent part does not contain any contain any qualifiers or
probability words. The propositions are expressed in simple canonical terms. Categorical
reasoning consists of the following four types of rules:

Qualitative Reasoning
This type of reasoning that combines the advantages of fuzzy reasoning and qualitative
reasoning.
Syllogistic Reasoning
A fuzzy syllogism is defined that is an inference schema that contains major premise, mi-
nor premise and conclusion as propositions containing fuzzy quantifiers. A basic fuzzy syllo-
gism in fuzzy logic is the intersection/product syllogism. The general form of syllogism is:

in which the first premise p(Q1) is a fuzzy proposition containing a fuzzy quantifier Q1, the
second premise q(Q2) is a fuzzy proposition containing a fuzzy quantifier Q2 and the conclusion
r(Q) is a fuzzy proposition containing a fuzzy quantifier Q. The following are the forms of
syllogism:
• Intersection / product syllogism: C=A ^ B, E=A, F=C ^ D.
• Chaining syllogism: C=B, E=A, F=D.
• Consequent conjunction syllogism: A=C=E, F=B ^D
 • Consequent disjunction syllogism: A=C=E, F=B V D
• Antecedent conjunction syllogism: B=D=F, E=A ^D
• Antecedent disjunction syllogism: B=D=F, E=A VD
Dispositional Reasoning
The antecedents contain the quantifier ‘usually’. The following are the rules of inference.

1.6 FUZZY INFERENCE SYSTEM (FIS)

Fuzzy inference is the process of cresting mapping from a given input to an output using
fuzzy logic. The mapping acts as a basis from which decisions could be made or patterns can be
created. The fuzzy inference systems are constructed with membership functions, fuzzy logic
operators, and if-then rules.

A fuzzy inference system (FIS) is a system that uses fuzzy set theory to map
inputs to outputs.
The system uses IF…THEN…ELSE rules along with fuzzy connectors “OR”, “AND” etc.
The input given to FIS may be crisp but the output will be a fuzzy set.

Fig 1.17 Fuzzy Inference system

The components of FIS are:
i) Rule base: This consists of number of IF…THEN rules.
ii) Data base: This consists of membership functions.
iii) Decision unit: This is responsible for choosing which operation is to be performed based
on the rule.
iv) Fuzzification unit: This unit converts crispy input to fuzzy quantities.
v) Defuzzification unit: This unit converts fuzzy input to crisp quantities.

A fuzzy inference system finds its application in automatic control, data classification,
decision analysis, expert systems, and computer vision. There are two types of fuzzy inference
systems:
• Mamdani Fuzzy Inference System
• SugenoFuzzy Inference System
Mamdani Fuzzy Inference System
• Mamdani fuzzy inference is the most commonly used inference method.
• This method was introduced by Mamdani and Assilian.
• This system was designed to control a steam engine and boiler combination by
formulating control rules obtained from experts.
• The Mamdani FIS obtains the output after the following six steps:
1. Determinine a set of fuzzy rules.
2. Fuzzify the inputs using the input membership functions.
3. Combine the fuzzified inputs according to the fuzzy rules to establish a rule strength.
4. Find the consequence of the rule by combining the rule strength and the output
membership function.
5. Combine the resulting consequences to get an output distribution.
6. Defuzzify the output distribution.

Fig 1.18 Example of Mamdani Fuzzy Inference System

The above example
• Fuzzifies a two input Mamdani FIS with two rules (i.e.) minimum and maximum by
finding the intersection of the crisp input value with the input membership function.
• It uses the minimum operator to compute the fuzzy “AND” for combining the two
fuzzified inputs to obtain a rule strength.
• It clips the output membership function exactly at the rule strength.
 • Then the FIS uses the maximum operator to compute the fuzzy “OR” for combining the
outputs obtained from the two rules.
The figure 1.19 a) and b) shows the input and the output of the given example

Advantages of Mamdani’s FIS:

• It is intuitive.
• It has widespread acceptance.
• It is well suited to human input.
• It is accepted when wide knowledge is demanded.
Disadvantages of Mamdani’s FIS:
• Computational cost is more.
Sugeno Fuzzy Inference Systems
This system is similar to the Mandami’s inference systems. The primary difference between
these systems is that the Sugeno’s model gives the output as membership functions with linear
or constant values.
• Consider the following inputs: input1=x and input2=y.
• Then the typical output value z=ax+by+c.
• The output level zi of each rule is weighted by the firing strength wi of the rule. Firing
strength is the degree to which the rule matches the inputs.
Advantages of the Sugeno Method
• It is computationally efficient.
• It can be used to model any inference system in which the output membership functions
are either linear or constant.
• It works well with linear techniques, optimization and adaptive techniques.
• It has guaranteed continuity of the output surface.
• It is well suited to mathematical analysis.

fuzzy set operations :

https://sunilwanjarisvpcet.wordpress.com/soft-computing/