Module 2.5
Module 2.5
and Inferences
Dr. Debaleena Datta
Topic Details
2
Fuzzy Implication Rule (FI)
• A fuzzy implication rule (a.k.a., fuzzy ‘if-then-else’ rule or fuzzy conditional
statement assumes the form:
“If x is A then y is B”
where x and y are two linguistic variables (i.e., variables whose values are
words or sentences in a natural or artificial language) defined by fuzzy sets A
and B on the universe of discourses X and Y, respectively
4
Fuzzy Implication Rule (FI) : Practical Example
• Suppose, P and T are two universes of discourse
• P= ‘Pressure’ = {1, 2, 3, 4}
• T= ‘Temperature’= {20, 25, 30, 35, 40, 45, 50}
• Let, the fuzzy implication rule “If temperature is high the pressure is low”
is defined as:
5
Fuzzy Implication Rule (FI) : Practical Example
• The linguistic variables ‘High Temperature’ and ‘Low Pressure’ is given as:
6
Interpretation of Fuzzy Implication Rules :
A Coupled with B
• This rule can mathematically be expressed as:
7
Interpretation of Fuzzy Implication Rules :
A Coupled with B
The most frequently used T-norm operators are:
8
Interpretation of Fuzzy Implication Rules :
A Entails B
9
Interpretation of Fuzzy Implication Rules :
A Entails B
10
Zadeh’s Max-min Rule
“If x is A then y is B”
11
Zadeh’s max-min rule : Example 1
Apply the Zadeh’s max-min rule to determine the FI rule (Rmm) for
“If x is A then y is B”
12
Zadeh’s max-min rule : Example 1
13
Zadeh’s max-min rule : Example 2
• Suppose, X ={a, b, c, d} and Y = {1, 2, 3, 4}
• Three Fuzzy sets A and B are given by:
A={(a, 0.0), (b, 0.8), (c, 0.6), (d, 1.0)}
B={(1, 0.2), (2, 1.0), (3, 0.8), (4, 0.0)}
C={(1, 0.0), (2, 0.4), (3, 1.0), (4, 0.8)}
Apply the Zadeh’s max-min rule to determine the FI rule (Rmm) for
“If x is A then y is B else y is C”
[Rmm =
The membership function Rmm of is given by:
14
Zadeh’s max-min rule : Example 2
15
Fuzzy Inferences
16
Fuzzy Inferences
• Two important inferring procedures used in fuzzy systems are:
17
Fuzzy Inferences
18
Fuzzy Inferences Example: GMP
19
Fuzzy Inferences Example: GMP
20
Fuzzy Inferences Example: GMP
21
Fuzzy Inferences Example: GMT
22
Fuzzy Inferences Example: GMT
23
Fuzzy Inferences: Practical Example
Consider the following proposition:
P: If service is good then customer is satisfied
Q: Service is very good
Customer is very satisfied
The universe of discourses are given as:
Service–rating = {a,b,e,d,e}
Satisfaction-grade= { 1,2,3,4, 5}
Both set having the elements in the ascending order.
The fuzzy sets for the propositions are given as:
Good-service= { (a, 1), (b, 0.8), (c,0.6) , (d,0.4) , (e,0.2)}
Satisfied = { (1,0.2) (2,0.4),(3,0.6),(4,0.8 ), (5,1)}
Very good-service= {(a,0.8), (b,0.6),(c, 0.4 ), (d, 0),( e, 0)}
Use appropriate FI rule to find the fuzzy set for ‘Very satisfied'.
24
Fuzzy Inferences: Practical Example
Consider another Proposition :
P: If service is good then customer is satisfied
Q1: Customer is not satisfied
Service is not good
If the fuzzy set is given as following:
Not-satisfied ={ (1,0),(2,0),(3,0.2),(4,0.4),(5,0.6)}
Use appropriate FI rule to find the fuzzy set for ' Not good-service '.
25
Fuzzy Inferences: Practical Example
For the first solution, we need to use the Generalized Modus Ponens(GMP)
rule. To find the FI Rule we first need to find out the following components:
Good-service X Satisfied =
26
Fuzzy Inferences: Practical Example
27
Fuzzy Inferences: Practical Example
28
Fuzzy Inferences: Practical Example
29
Fuzzy Inferences: Practical Example
For the second solution, we need to use the Generalized Modus
Tollens(GMT) rule. We have already found Rmm. Thus,
30
Fuzzy Inferences: Practical Example
31