Geoscience Reference
In-Depth Information
3.2 Compositional Rule of Inference
Practically we need to interpret verbal values of sets A, B mathematically and de
ne
the rule of fuzzy relation R between variables X, Y. We use the compositional rule of
inference for assignment value B
′
of variable Y, which corresponds with value A
′
of
variable X.
We can get term, where the set B
′
is the sup-min composition of the fuzzy set A
′
and the fuzzy relation R, written as B
0
¼
A
0
R with the membership [
6
]
l
B
0
ð
y
Þ¼
sup
x
2
X
min
ð
l
A
0
ð
x
Þ
;
l
R
ð
x
;
y
ÞÞ
standard intersection
or generally
l
B
0
ð
y
Þ¼
sup
x
2
X
T
ð
l
A
0
ð
x
Þ
;
l
R
ð
x
;
y
ÞÞ
union based on t-norm T
Rule
(X, Y)isR(A, B)
X is A
′
Observing
compositional rule of inference on t-norm T
, B
0
¼
A
0
T
R
ð
A
;
B
Þ
Conclusion
Y is B
′
We have to keep generalized modus ponens during relational reasoning, too, i.e.
A
T
R
ð
A
;
B
Þ¼
B.
The fuzzy relations can be modelled by a logical implication or by a cartesian
product T
based on t-norm. We con
ne to the second possibility and we get:
l
R
ð
A
;
B
Þ
ð
x
;
y
Þ¼
T
ð
l
A
ð
x
Þ
;
l
B
ð
y
ÞÞ
l
B
0
ð
y
Þ¼
sup
x
2
ð
l
A
0
ð
x
Þ
;
T
l
A
ð
x
Þ
;
l
B
ð
y
ÞÞÞ
min
X
We can generalize the properties to t-norm T:
ðÞ
;
T
l
A
ðÞ
;
l
B
ðÞ
l
B
0
ðÞ¼
sup
x
2
X
T l
A
0
ð
ð
Þ
Þ
If we choose T
¼
T
¼
T
M
we get:
l
B
0
ðÞ¼
sup
x
2X
min
ð
l
A
0
ðÞ
;
min
ð
l
A
ðÞ
;
l
B
ðÞ
Þ
Þ
Mamdani's method
For T
¼
T
M
and T
¼
T
P
, it is: [
5
].
l
B
0
ðÞ¼
sup
x
2X
min
ð
l
A
0
ðÞ
;
l
A
ðÞ
l
B
ðÞ
Þ
Larsen's method
Search WWH ::
Custom Search