Topic

Lecture 01: Introduction to Fuzzy Sets

Concepts Covered:

 Classical Set/Crisp Set

 Properties of Classical Set/Crisp Set

 Fuzzy Set

 Representation of Fuzzy Set

Classical Set/Crisp Set
(A)
• Universal Set/Universe of Discourse (X): A set consisting of all
possible elements
Ex: All technical universities in the world
• Classical or Crisp Set is a set with fixed and
well-defined boundary
• Example: A set of technical universities having at least five
departments each
Representation of Crisp

• A={a1,a2,……,an}
Sets

• A={x|P(x)}, P: property
• Using characteristic function

1, if x belongs to A,
μA(x)=
0, if x does not belong to A.
Notations Used in Set Theory
•  : Empty/Null set
• x  A : Element x of the Universal set X belongs to set A
• x  A : x does not belong to set A
• A  B : set A is a subset of set B
• A  B : set A is a superset of set B
• A B : A and B are equal
• A B : A and B are not equal
• A  B : A is a proper subset of B
• A  B : A is a proper superset of B
• A : Cardinality of set A is defined as the total number of
elements present in that set
• p(A) : Power set of A is the maximum number of subsets
including the null that can be constructed from a set A

Note: p A  2 A
Crisp Set Operations
• Difference:
It is known as relative complement of set B with respect to set A

Absolute complement: 𝑨= 𝑨 𝑪 = 𝑿 − 𝑨={𝒙∨𝒙 ∈ 𝑿 𝒂𝒏𝒅 𝒙 ∉ 𝑨 }

• Intersection:
• Union:
Properties of Crisp Sets

1. Law of involution:
2. Law of Commutativity:
3. Associativity: ;

5. Laws of Tautology: ;
4. Distributivity:

6. Laws of Absorption:
7. Laws of Identity:
8. De Morgan’s Laws:
9. Law of contradiction:
10. Law of excluded middle:
Fuzzy Sets

• Sets with imprecise/vague boundaries

• Introduced by Prof. L.A. Zadeh, University of California, USA, in 1965
• Potential tool for handling imprecision and uncertainties
• Fuzzy set is a more general concept of the classical set
Representation of a Fuzzy Set

Ax  x,  A x , x  X 

Note:
Probability: Frequency of likelihood that an element is in a class
Membership: Similarity of an element to a class
 Types of Fuzzy sets

1. Discrete Fuzzy set

n
Ax    A xi / xi ,
i 1
n: Number of elements present in the set

2. Continuous Fuzzy set

Ax    A x / x
X
Convex vs. Non-Convex Membership Function Distribution
A fuzzy set A(x) will be convex, if
 A x1  1   x2  min  A x1 ,  A x2 
Where 0.0 ≤ λ ≤ 1.0
Various Types of Membership Function Distributions

1. Triangular Membership

  x a c x 
 triangle max  min  ,  ,0 
  b a c b 
2. Trapezoidal Membership

  x a d  x 
trapezoidal max  min  ,1, ,0
  b a d  c  
3. Gaussian Membership

1
 Gaussian  2
1  x m 
 
2  
e
4. Bell-shaped Membership Function

1
 Bell  shaped  2b
x c
1
a
5. Sigmoid Membership

1
 Sigmoid   a x  b 
1 e
Summary:

Classical Set/Crisp Set has been defined

Properties of Classical Set/Crisp Set has been explained

Fuzzy Set has been defined

Deals with representation of Fuzzy Set

Concepts Covered:

 A few terms of Fuzzy Sets

 Standard Operations in Fuzzy Sets

 Properties of Fuzzy Sets

 Numerical Example
Triangular Membership: Determine μ,
corresponding to x=8.0

1.0

0.0
a=2 b=6 8 c=10
x
𝝁𝒕𝒓𝒊𝒂𝒏𝒈𝒍𝒆 =𝒎𝒂𝒙 [ 𝒎𝒊𝒏 ( 𝒙−𝒂 𝒄− 𝒙
,
𝒃−𝒂 𝒄−𝒃 )
,𝟎 ]

We put, x=8.0

[ ( ) ]
𝝁𝒕𝒓𝒊𝒂𝒏𝒈𝒍𝒆 =𝒎𝒂𝒙 𝒎𝒊𝒏
𝟑 𝟏
,
𝟐 𝟐
𝟏
, 𝟎 = =𝟎 . 𝟓
𝟐
Trapezoidal Membership
•Determine μ corresponding to x = 3.5
𝜇 𝒕𝒓𝒂𝒑𝒆𝒛𝒐𝒊𝒅𝒂𝒍max  min x a
,1,
d x 
 ,0 
  b a d c 

  x  2 10  x  
max  min  ,1,  ,0 
  4 2 10  8  

  x  2 10  x  
max  min  ,1,  ,0 
  2 2  
•We put x = 3.5

𝜇 𝒕𝒓𝒂𝒑𝒆𝒛𝒐𝒊𝒅𝒂𝒍max  min 1.5 ,1, 6.5  ,0

  2 2  

max0.75,0

0.75
Gaussian Membership:
Determine corresponding to

1.0

0.0
𝒎
𝒙
 𝟏
𝝁 𝑮𝒂𝒖𝒔𝒔𝒊𝒂𝒏 = 𝟐

𝒆
𝟐 (
𝟏 𝒙 −𝒎
𝝈 )

𝑻𝒂𝒌𝒆𝒎=𝟏𝟎 . 𝟎 𝒂𝒏𝒅 𝝈 =𝟑 .𝟎
𝟏
𝝁 𝑮𝒂𝒖𝒔𝒔𝒊𝒂𝒏 = 𝟐

𝒆
𝟏
𝟐 ( 𝒙 −𝟏𝟎 .𝟎
𝟑 .𝟎 )
We put

𝟏
∴ 𝝁 𝑮𝒂𝒖𝒔𝒔𝒊𝒂𝒏= 𝟐 =𝟎 .𝟗𝟒𝟓𝟗
𝒆
𝟐(
𝟏 𝟗 .𝟎 −𝟏𝟎 .𝟎
𝟑.𝟎 )
Bell-shaped Membership function: Determine μ
corresponding to

1.0

0.0
c
x
 𝟏
𝝁 𝑩𝒆𝒍𝒍 − 𝒔𝒉𝒂𝒑𝒆𝒅 =
| |
𝟐𝒃
𝒙−𝒄
𝟏+
𝒂

Take c=10.0, a=2.0, b=3.0

𝟏
𝝁 𝑩𝒆𝒍𝒍 − 𝒔𝒉𝒂𝒑𝒆𝒅 =
| |
𝟔
𝒙 − 𝟏𝟎
𝟏+
𝟐
We put x=8.0
𝟏
𝝁 𝑩𝒆𝒍𝒍 − 𝒔𝒉𝒂𝒑𝒆𝒅 = =𝟎 . 𝟓
| |
𝟔
𝟖 − 𝟏𝟎
𝟏+
𝟐
Sigmoid Membership Function:

Determine µ corresponding to x = 8.0

1  e a( x b)
1
 sigmoid


Take b = 6.0; a = 2

1e
sigmoid  12( x 6.0)

we put x = 8.0

0.98
1e 1e
1 1
 sigmoid
  22.0
 4
Difference Between Crisp and Fuzzy
Sets
 A Few Definitions in Fuzzy Sets
• α-cut of a fuzzy set

A set consisting of elements x of the Universal set X, whose membership

values are either greater than or equal to the value of α.
• Strong α-cut of a Fuzzy Set

+ ¿ ( 𝒙 ) = { 𝒙 |𝝁 ( 𝒙 )> 𝜶 } ¿
𝜶 𝝁𝑨
𝑨
 Numerical Example
The membership function distribution of a fuzzy set is assumed to follow a
Gaussian distribution with mean m = 100 and standard deviation σ =20 .
Determine 0.6 – cut of this distribution.
Solution:
Gaussian distribution : 1
 1  x m 

e
2

 
2  

where m : Mean ; σ : Standard deviation

By substituting the values of µ = 0.6, m = 100, σ =20 and
taking log (ln) on both sides, we get
 0.6 
1
1  x 100 

e
2

 
2  20 

1  x 100 

 e
2

0.6

2  20 
 1


By taking ln

 1  x 100  2 
ln e2  20   ln1.6667 
  Figure : 0.6-cut of a fuzzy set.

 x ( 79.7846 ,120.2153)
 
• Support of a Fuzzy Set A(x)

It is defined as the set of all x є X, such that μA(x) > 0

supp(A)

Note: Support of a fuzzy set is nothing but its Strong 0-cut

• Scalar Cardinality of a Fuzzy Set A(x)

| 𝑨 ( 𝒙 )|= ∑ 𝝁 𝑨 ( 𝒙 )
𝒙∈ 𝑿
 Numerical Example

Let us consider a fuzzy set A  x  as follows:

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 

 
Scalar Cardinality A x 0.1  0.2  0.3  0.4 1.0
• Core of a Fuzzy Set A(x)
It is nothing but its 1-cut

• Height of a Fuzzy Set A(x)

It is defined as the largest of membership values of the elements
contained in that set.
• Normal Fuzzy Set
For a normal fuzzy set, h(A) = 1.0

• Sub-normal Fuzzy Set

For a sub-normal fuzzy set, h(A) < 1.0
 Some Standard Operations in Fuzzy
Sets
• Proper Subset of a Fuzzy Set

𝑨 ( 𝒙 ) ⊂ 𝑩 ( 𝒙 ) , 𝒊𝒇 𝝁 𝑨 ( 𝒙 ) <𝝁 𝑩 ( 𝒙 )
Numerical Example

Let us consider the two fuzzy sets, as follows:

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 
 
B x   x ,0.5  , x ,0.7  , x ,0.8  , x ,0.9 
1 2 3 4

As for all x  X,  A  x   B  x  ,
A x  B x , that is , A x is the proper subset of B x
 Some Standard Operations in Fuzzy
Sets (contd.)

• Equal fuzzy sets

𝑨 ( 𝒙 )= 𝑩 ( 𝒙 ) ,𝒊𝒇 𝝁 𝑨 ( 𝒙 )=𝝁 𝑩 ( 𝒙 )
Numerical Example

Let us consider the two fuzzy sets, as follows:

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 
 
B x   x ,0.5  , x ,0.7  , x ,0.8  , x ,0.9 
1 2 3 4

As for all x  X,  A  x  B  x , A x B x

• Complement of a Fuzzy Set

𝑨 ( 𝒙 ) =𝟏 − 𝑨 ( 𝒙 )
Numerical Example

Let us consider a fuzzy set A  x  as follows:

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 

 
Complement A x   x1 ,0.9  , x2 ,0.8  , x3 ,0.7  , x4 ,0.6 
 
• Intersection of Fuzzy Sets

Intersection of two fuzzy sets A(x) and B(x) is denoted by

and its membership values are determined as follows :

𝝁( 𝑨∩ 𝑩 ) ( 𝒙 )=𝒎𝒊𝒏 {𝝁 𝑨 ( 𝒙 ) ,𝝁 𝑩 ( 𝒙 ) }
 x

Note: Intersection is analogous to logical AND operation

Numerical Example
Let us consider the two fuzzy sets as follows:
A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 
 
B x   x ,0.5  , x ,0.7  , x ,0.8  , x ,0.9 
1 2 3 4

Now,  AB  x1  min  A  x1  ,B  x1  min 0.1,0.5 0.1

 
Similarly,  AB  x2  min 0.2 ,0.7 0.2
AB  x3  min 0.3 ,0.8 0.3
AB  x4  min 0.4,0.9 0.4
• Union of Fuzzy Sets

Union of two fuzzy sets A(x) and B(x) is represented by:

and its membership value is determined as follows:

𝝁( 𝑨∪ 𝑩 ) ( 𝒙 )= 𝒎𝒂𝒙 {𝝁 𝑨 ( 𝒙 ) , 𝝁 𝑩 ( 𝒙 ) }
Note: Union is analogous to logical OR operation
Numerical Example
Let us consider the following two fuzzy sets:
A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 
 
B x   x ,0.5  , x ,0.7  , x ,0.8  , x ,0.9 
1 2 3 4

Now,  AB  x1  max  A  x1  ,B  x1  max 0.1,0.5 0.5

 
Similarly,  AB  x2  max 0.2 ,0.7 0.7
AB  x3  max 0.3 ,0.8 0.8
AB  x4  max 0.4 ,0.9 0.9
• Algebraic product of Fuzzy Sets

𝑨 ( 𝒙 ) . 𝑩 ( 𝒙 )={ ( 𝒙 , 𝝁 𝑨 ( 𝒙 ) .𝝁 𝑩 ( 𝒙 ) ) , 𝒙 ∈ 𝑿 }
Numerical Example

Let us consider the following two fuzzy sets:

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 
 
B x   x ,0.5  , x ,0.7  , x ,0.8  , x ,0.9 
1 2 3 4

A x.B x   x1 ,0.05  , x2 ,0.14  , x3 ,0.24  , x4 ,0.36 

 
• Multiplication of a Fuzzy Set by a Crisp
Number

𝒅 . 𝑨 ( 𝒙 )={ ( 𝒙 , 𝒅 × 𝝁 𝑨 ( 𝒙 ) ) , 𝒙 ∈ 𝑿 }
Numerical Example

Let us consider a fuzzy set

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4  and a crisp number d  0.2

 
d.A x   x1 ,0.02  , x2 ,0.04  , x3 ,0.06  , x4 ,0.08 
 
• Power of a Fuzzy Set

AP(x): p-th power of a fuzzy set A(x) such that

𝒑
𝝁 𝑨 ( 𝒙 ) = {𝝁 𝑨 ( 𝒙 ) } , 𝒙 ∈ 𝑿
𝒑

Concentration: p=2
Dilation: p=1/2
Numerical Example

Let us consider a fuzzy set

A x   x1 ,0.1 , x2 ,0.2  ,x3 ,0.3  ,x4 ,0.4  and power p  2

 
A2  x   x1 ,0.01 , x2 ,0.04  ,x3 ,0.09  ,x4 ,0.16 
 
• Algebraic Sum of two Fuzzy Sets A(x) and B(x)

𝑨 ( 𝒙 ) + 𝑩 ( 𝒙 )={( 𝒙 , 𝝁 𝑨+ 𝑩 ( 𝒙 ) ) , 𝒙 ∈ 𝑿 }

where

𝝁 𝑨+𝑩 ( 𝒙 )=𝝁 𝑨 ( 𝒙 ) +𝝁 𝑩 ( 𝒙 ) − 𝝁 𝑨 ( 𝒙 ) . 𝝁 𝑩 ( 𝒙 )
Numerical Example

Let us consider the following two fuzzy sets:

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 
 
B x   x ,0.5  , x ,0.7  , x ,0.8  , x ,0.9 
1 2 3 4

 A x  B x   x1 ,0.55  , x2 ,0.76  , x3 ,0.86  , x4 ,0.94 

 
• Bounded Sum of two Fuzzy Sets

𝑨 ( 𝒙 ) ⨁ 𝑩 ( 𝒙 )={( 𝒙 , 𝝁 𝑨 ⨁ 𝑩 ( 𝒙 ) ), 𝒙 ∈ 𝑿 }

where

𝝁 𝑨⨁ 𝑩 ( 𝒙 )=𝒎𝒊𝒏 {𝟏 , 𝝁 𝑨 ( 𝒙 ) +𝝁 𝑩 ( 𝒙 ) }
Numerical Example

Let us consider the following two fuzzy sets:

A x   x1 ,0.1 , x2 ,0.2  , x3 ,0.3  , x4 ,0.4 
 
B x   x ,0.5  , x ,0.7  , x ,0.8  , x ,0.9 
1 2 3 4

 A x  B x   x1 ,0.6  , x2 ,0.9  , x3 ,1.0  , x4 ,1.0 

 
• Algebraic Difference of two Fuzzy Sets

𝑨 ( 𝒙 ) − 𝑩 ( 𝒙 )={( 𝒙 ,𝝁 𝑨− 𝑩 ( 𝒙 )) , 𝒙 ∈ 𝑿 }

where

𝝁 𝑨− 𝑩 ( 𝒙 ) =𝝁 𝑨∩ 𝑩 (𝒙)
Numerical Example

•Let us consider the following two fuzzy sets:

A( x) {( x1 , 0.1), ( x2 , 0.2), ( x3 , 0.3), ( x4 , 0.4)}

B ( x) {( x1 , 0.5), ( x2 , 0.7), ( x3 , 0.8), ( x4 , 0.9)}
Now, B ( x ) {( x1 , 0.5), ( x2 , 0.3), ( x3 , 0.2), ( x4 , 0.1)}

 A( x)  B ( x) {( x1 , 0.1), ( x2 , 0.2), ( x3 , 0.2), ( x4 , 0.1)}

• Bounded Difference of two Fuzzy Sets

𝑨 ( 𝒙 ) 𝚯 𝑩( 𝒙 )={( 𝒙 , 𝝁 𝑨 𝚯 𝑩 ( 𝒙 )) , 𝒙 ∈ 𝑿 }

where

𝝁 𝑨 𝚯 𝑩 ( 𝒙 )=𝒎𝒂𝒙 {𝟎 , 𝝁 𝑨 ( 𝒙 ) +𝝁 𝑩 ( 𝒙 ) −𝟏 }
Numerical Example

•Let us consider the following two fuzzy sets:

A( x) {( x1 , 0.1), ( x2 , 0.2), ( x3 , 0.3), ( x4 , 0.4)}

B ( x) {( x1 , 0.5), ( x2 , 0.7), ( x3 , 0.8), ( x4 , 0.9)}

A( x)B ( x) {( x1 , 0.0), ( x2 , 0.0), ( x3 , 0.1), ( x4 , 0.3)}

• Cartesian product of two Fuzzy Sets

Two fuzzy sets A(x) defined in X

and B(y) defined in Y
Cartesian product of two fuzzy sets is denoted by A(x)×B(y),
𝑨× 𝑩 ( 𝒙 , 𝒚 ) =𝒎𝒊𝒏 {𝝁 𝑨 ( 𝒙 ) , 𝝁 𝑩 ( 𝒚 ) }
𝝁that
such
Numerical Example
•Let us consider the following two fuzzy sets:

A( x) {( x1 , 0.2), ( x2 , 0.3), ( x3 , 0.5), ( x4 , 0.6)}

B ( y ) {( y1 , 0.8), ( y2 , 0.6), ( y3 , 0.3)}

min(  A ( x1 ),  B ( y1 )) min(0.2, 0.8) 0.2

min(  A ( x1 ),  B ( y2 )) min(0.2, 0.6) 0.2
min(  A ( x1 ),  B ( y3 )) min(0.2, 0.3) 0.2
min(  A ( x2 ),  B ( y1 )) min(0.3, 0.8) 0.3
min(  A ( x2 ),  B ( y2 )) min(0.3, 0.6) 0.3
min(  A ( x2 ),  B ( y3 )) min(0.3, 0.3) 0.3
min(  A ( x3 ),  B ( y1 )) min(0.5, 0.8) 0.5
min(  A ( x3 ),  B ( y2 )) min(0.5, 0.6) 0.5
min(  A ( x3 ),  B ( y3 )) min(0.5, 0.3) 0.3
min(  A ( x4 ),  B ( y1 )) min(0.6, 0.8) 0.6
min(  A ( x4 ),  B ( y2 )) min(0.6, 0.6) 0.6
min(  A ( x4 ),  B ( y3 )) min(0.6, 0.3) 0.3

 0.2 0.2 0.2 

 0.3 0.3 0.3
 A B 
 0.5 0.5 0.3
 
 0.6 0.6 0.3 
 Composition of fuzzy relations
Let A = [aij] and B = [bjk] be two fuzzy relations expressed in the matrix form.

Composition of these two fuzzy relations, that is, C is represented as follows:

C=A о B

In matrix form
[cik] = [aij] о [bjk]
Where
cik =max[min(aij, bjk)]
Numerical Example
•Let us consider the following two Fuzzy relations:
 0.2 0.3
 
A  a ij  
 0.5 0.7 

 0.3 0.6 0.7 

 
B  b jk 
0.6 
 0.1 0.8

•Elements of c ik  matrix can be determined as follows:

c11 maxmin( a11 , b11 ), min( a12 , b21 )

maxmin( 0.2,0.3), min( 0.3,0.1)

max0.2,0.1

0.2
c12 maxmin( a11 , b12 ), min( a12 , b22 )

maxmin( 0.2,0.6), min( 0.3,0.8)

max0.2,0.3

0.3
c13 maxmin( a11 , b13 ), min( a12 , b23 )

maxmin( 0.2,0.7 ), min( 0.3,0.6)

max0.2,0.3

0.3
c21 maxmin( a 21 , b11 ), min( a 22 , b21 )

maxmin( 0.5,0.3), min( 0.7,0.1)

max0.3,0.1

 0 .3
c22 maxmin( a 21 , b12 ), min( a 22 , b22 )

maxmin( 0.5,0.6), min( 0.7,0.8)

max0.5,0.7

0.7
c23 maxmin( a 21 , b13 ), min( a 22 , b23 )

maxmin( 0.5,0.7 ), min( 0.7,0.6)

max0.5,0.6

 0 .6
  0.2 0.3 0.3
 C 
 0.3 0.7 0.6
 Properties of Fuzzy Set
Fuzzy sets follow the properties of crisp sets except the following two:

• Law of excluded middle

In crisp set,

In fuzzy set,
• Law of contradiction

In crisp set, O

In fuzzy set, O
Summary:

• A few terms related to Fuzzy Sets have been defined

• Some standard Operations in Fuzzy Sets have been

explained

• Properties of Fuzzy Sets have been explained