Fuzzy Membership Functions

Fuzzy Operations
Fuzzy Union
F U i
Fuzzy Intersection
Fuzzy Complement
 Some info for LAB
• Work on an m-file (open m-file for each task,
write your programme, save the file (e.g.,
lab2task1), then execute the file. Now, this
fil h
file has bbecome a ffunction
ti iin MATLAB)
MATLAB). ((see
the first week’s slides - Week 1).

• Use help <function

<function>
> (e.g., help newfis) if
you don’t’ know how to use the ffunction. It
gives you information about how to use the
f
function
ti and d what
h t parameters
t it requires
i
Fuzzy Membership Functions
• One of the keyy issues in all fuzzyy sets is how to
determine fuzzy membership functions
• The membership function fully defines the fuzzy
set
• A membership function provides a measure of
the degree of similarity of an element to a fuzzy
set
e be s p functions
• Membership u ct o s ca
can ta
take
ea anyy form,
o , but
there are some common examples that appear
in real applications
• Membership functions can
– either be chosen by the user arbitrarily, based
on the user’s experience
p ((MF chosen byy two
users could be different depending upon their
experiences, perspectives, etc.)
– Or
O beb ddesigned
i d using
i machine hi llearning
i
methods (e.g., artificial neural networks,
genetic algorithms, etc.)
• There are different shapes of membership
functions; triangular, trapezoidal,
piecewise-linear, Gaussian, bell-shaped,
etc.
• Triangular membership function
– a, b and c represent the x coordinates of the three
vertices of µA(x) in a fuzzy set A (a: lower boundary
and c: upper boundary where membership degree is
zero, b: the centre where membership degree is 1)

µA(x)
1
⎧0 if x ≤ a ⎫
⎪x −a ⎪
⎪ if a ≤ x ≤ b ⎪
⎪ ⎪
μ A ( x) = ⎨ b − a ⎬
⎪c − x if b ≤ x ≤ c ⎪
⎪c − b ⎪
⎪
⎩0 if x ≥ c ⎪
⎭
0
a b c x
• Gaussian membership function
⎡ 1 x−c m
⎤
μ A ( x, c, s, m) = exp ⎢− ⎥
⎢⎣ 2 s ⎥⎦
– c: centre
– s: width
– m: fuzzification factor (e.g., m
m=2)
2)
µA(x) 1

0.9
c=5
0.8

0.7

0.6
s=2
0.5

0.4 m=22
0.3

0.2

0.1

0
0 1 2 3 4 5 6 7 8 9 10
x
 1

0.9

08
0.8

0.7 c=5
0.6

0.5

04
0.4
s=0.5
0.3

0.2

0.1
m=2
0
0 1 2 3 4 5 6 7 8 9 10

0.9

0.8

0.7
c=5
0.6

0.5

0.4
s=5
0.3

0.2
m=2
0.1

0
0 1 2 3 4 5 6 7 8 9 10
 1

09
0.9

0.8
c=5
0.7

0.6
s=2
0.5

0.4
m=0.2
0.3
0 1 2 3 4 5 6 7 8 9 10

09
0.9

0.8 c=5
0.7

0.6

05
0.5
s=5
0.4

0.3

0.2
m=5
01
0.1

0
0 1 2 3 4 5 6 7 8 9 10
 Fuzzy Operations
(Fuzzy Union
Union, Intersection
Intersection, and
Complement)
• F
Fuzzy llogic
i begins
b i by b barrowing
b i notions
ti from
f crisp
i logic,
l i just
j t as
fuzzy set theory borrows from crisp set theory. As in our extension
of crisp set theory to fuzzy set theory, our extension of crisp logic to
fuzzyy logic
g is made byy replacing
p g membershipp functions of crispp
logic with fuzzy membership functions [J.M. Mendel, Uncertain Rule-
Based Fuzzy Logic Systems, 2001]
• In Fuzzy Logic, intersection, union and complement are defined in
terms of their membership functions
• This section concentrates on providing enough of a theoretical base
for you to be able to implement computer systems that use fuzzy
logic
g
• Fuzzy intersection and union correspond to ‘AND’ and ‘OR’,
respectively, in classic/crisp/Boolean logic
• These two operators will become important later as they are the
b ildi blocks
building bl k forf us to beb able
bl to compute with
i h fuzzy
f if-then
if h rules
l
 Classic/Crisp/Boolean Logic
• Logical AND (∩) • Logical OR (U)
Truth Table Truth Table
A B A∩B A B AUB
0 0 0 0 0 0
0 1 0 0 1 1
1 0 0 1 0 1
1 1 1 1 1 1

A
B A
B

Crisp Intersection Crisp Union

Fuzzy Union
• The union (OR) is calculated using t-conorms
• tt-conorm
conorm operator is a function s(.,.)
• Its features are
– s(1,1) = 1, s(a,0) = s(0,a) = a (boundary)
– s(a,b) ≤ s(c,d) if a ≤ c and b ≤ d (monotonicity)
– s(a,b) = s(b,a) (commutativity)
– s(a,s(b,c))
( (b )) = s(s(a,b),c)
( ( b) ) (associativity)
( i ti it )
• The most commonly used method for fuzzy union is to
take the maximum.
maximum That is,
is given two fuzzy sets A and B
with membership functions µA(x) and µB(x)

μ AUB ( x) = max((μ A ( x),

) μ B ( x))
Fuzzy Intersection
• The intersection (AND) is calculated using t-norms.
• tt-norm
norm operator is a function t(.,.)
• Its features
– t(0,0) = 0, t(a,1) = t(1,a) = a (boundary)
– t(a,b) ≤ t(c,d) if a ≤ c and b ≤ d (monotonicity)
– t(a,b) = t(b,a) (commutativity)
– t(a t(b,c))
t(a, t(b c)) = t(t(a
t(t(a,b),c)
b) c) (associativity)
• The most commonly adopted t-norm is the minimum.
That is,, given
g two fuzzyy sets A and B with membership
p
functions µA(x) and µB(x)
μ A∩ B ( x) = min(μ A ( x),
) μ B ( x))
Fuzzy Complement
• To be able to develop fuzzy systems we also have
to deal with NOT or complement.
p
• This is the same in fuzzy logic as for Boolean
logic
• For a fuzzy set A, A denotes the fuzzy
complement of A
• Membership function for fuzzy complement is

μ ( x) = 1 − μ A ( x)
A
Example 1:
Suppose we have the following (discrete) fuzzy sets:
A = 0.4/1+0.6/2+0.7/3+0.8/4
B = 0.3/1+0.65/2+0.4/3+0.1/4

The union of the fuzzy sets A and B

= 0.4/1+0.65/2+0.7/3+0.8/4
0 4/1+0 65/2+0 7/3+0 8/4
The intersection of the fuzzy sets A and B
= 0.3/1+0.6/2+0.4/3+0.1/4
0 3/1+0 6/2+0 4/3+0 1/4
The complement of the fuzzy set A
= 0.6/1+0.4/2+0.3/3+0.2/4
Example 1: (cont.)
Let’s show the fuzzy sets A and B graphically

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0 0
1 2 3 4 5 1 2 3 4 5

A B
Example 2 (2003 exam question)

Given two fuzzy sets A and B

a. Represent A and B fuzzy sets graphically

b. Calculate the of union of the set A and set B
c. Calculate the intersection of the set A and set B
d. Calculate the complement of the union of A and B

A = 0.0 / − 2 + 0.3 / − 1 + 0.6 / 0 + 1.0 / 1 + 0.6 / 2 + 0.3 / 3 + 0.0 / 4

B = 0.1 / − 2 + 0.4 / − 1 + 0.7 / 0 + 1.0 / 1 + 0.5 / 2 + 0.2 / 3 + 0.0 / 4

Example 2 (cont)

a
Example
p 2 ((cont))
b
Union = max (A,
(A B) = 0.1 / − 2 + 0.4 / − 1 + 0.7 / 0 + 1.0 / 1 + 0.6 / 2 + 0.3 / 3 + 0.0 / 4

c
Intersection = min (A,
(A B) = 0.0 / − 2 + 0.3 / − 1 + 0.6 / 0 + 1.0 / 1 + 0.5 / 2 + 0.2 / 3 + 0.0 / 4

d
Complement of (b) = 1 - max(A, B) = 0.9 / − 2 + 0.6 / − 1 + 0.3 / 0 + 0.0 / 1 + 0.4 / 2 + 0.7 / 3 + 1.0 / 4
Example 3: Graphical representation of the Fuzzy
operations (taken from J
J.M.
M Mendel
Mendel, Uncertain
Rule-Based Fuzzy Logic Systems, 2001)
Consider the fuzzy sets A = damping ratio x considerably
larger than 0.5, and B = damping ratio x approximately
equal to 0.707. Note that damping ratio is a positive real
number, i.e., its universe of discourse, X, is the positive
real numbers 0 ≤ x ≤ 1
Consequently, A = {( x, μ A ( x)) xεX } and B = {( x, μ B ( x)) xεX }
where, example, µA(x) and µB(x) are specified,
where for example specified as:
⎧0 if 0 ≤ x ≤ 0.5⎫
⎪ ⎪ μ B ( x) =
1
0 ≤ x ≤1
μ A ( x) = ⎨ 1 ⎬
if 0.5 < x ≤ 1 ⎪ 1 + ( x − 0.707) 4
⎪1 + ( x − 0.5) − 2
⎩ ⎭
Example 3: (cont.)
( ) µA(x),
Figure (a):
Fi ( ) µB(x)
( ) (b) µAUB(x)
Figure (b):
Fi ( )
Figure (c): µA∩B(x) Figure (d): µB(x), µB(x)
Example 3: (cont.)
• This example demonstrates that for fuzzy sets
sets, the
Law of Excluded Middle and Concentration are
broken,, i.e.,, for fuzzy
y sets A and B:
− −
A ∪ A ≠ X and A ∩ A ≠ O/
• In fact, one of the ways
y to describe the difference
between crisp set theory and fuzzy set theory is to
explain that these two laws do not hold in fuzzy set
theory