Information Technology Reference
In-Depth Information
ε
′
=
x
∈
R
ε
′
x
=
′
′
′
′
′
max{
|
( )
1}
M
=
max{
x
,
y
,
α β ε
,
,
}
ij
ij
′
x
ij
x
=
,
i
=
1
2
?
,
n
;
j
=
1
2
?
p
ij
M
′
y
ij
y
=
,
i
=
1
2
?
,
n
;
j
=
1
2
?
q
ij
M
′
′
′
α
β
γ
α
=
,
β
=
,
γ
=
M
M
M
x
ij
,
ŋ
,
Ȳ
,
ō
are closed positive fuzzy numbers with boundaries or zero
fuzzy numbers, their regular points are in the interval of 0,1.
ij
,
y
Definitions 12.6
The general support rate S about the general fuzzy
association rules in general sense “if
X
is
D
then
Y
is
E
” is defined as
n
p
q
Ã
∏
∏
[
a x
(
)
×
a y
(
)]
ij
ij
x
,
x
≥
ε
′
;
Ê
=
Ì
!
i
=
1
j
=
1
j
=
1
S
=
a x
( )
,
(12.3)
n
θ
others.
Definitions 12.7
The general confidence C about about the general fuzzy
association rules “if
X
is
D
then
Y
is
E
” isdefined as
x
,
x
≥
ε
′
Ê
=
Ì
!
S
ij
ij
C
=
a x
(
)
,
(12.4)
p
ij
1
n
θ
others
Ã
∏
[
a x
(
)]
ij
n
i
=
1
j
=
1
1
n
p
!
, then
Ã
∏
=
!
, and the rule will not be adopted.
S
When
[
a x
(
)]
=
θ
ij
n
i
=
1
j
=
1
1
n
p
!
it is easy to find that S in definition 12.6 and C in
Ã
∏
Supposed
[
a x
(
)]
≠
θ
ij
n
i
=
1
j
=
1
