Databases Reference
In-Depth Information
Tabl e 3 .
The fuzzy sets transformed by using chromosome
C
1
in
P opulation
3
TID Transformed Fuzzy set
T
1
0
.
6
cookies.LM
cookies.MH
0
.
14
+
cookies.MH
0
.
86
T
2
cookies.MH
0
.
86
T
3
cookies.LM
1
T
4
T
5
Null
T
6
Null
Tabl e 4 .
The counts of the fuzzy regions for item
cookies
when using
C
1
P opulation
3
Count
cookies.Low
0.00
cookies.LowMiddle
1.60
cookies.MiddleHigh
1.86
cookies.High
0.00
STEP 3.2: The scalar cardinality of each fuzzy region in the transactions
is calculated as the count value. Take the fuzzy region
cookies.LM
as an
example. Its scalar cardinality = (0.6 + 0.0 + 0.0 + 1 + 0.0 + 0.0) = 1.6. The
counts of the fuzzy regions for item
cookies
using
C
1
are shown in Table 4.
STEP 3.3: The count of any fuzzy region is checked against the predefined
minimum support value
α
. Assume in this example,
α
is set at 0.25. Since
both the count value of
cookies.LowMiddle
and
cookies.MiddleHigh
is larger
than 0.25*6 (= 1.5),
cookies.LowMiddle
and
cookies.MiddleHigh
is then put
in
L
1
.
STEP 3.4: Two large 1-itemset,
cookies.LowMiddle
and
cookies.Middle
High
, are derived from the membership functions of
C
1
in
Population
3
.
The fuzzy support of
cookies.LowMiddle
and
cookies.MiddleHigh
are
1.6/6 (= 0.266) and 1.86/6 (= 0.31). The suitability of
C
1
is calculated
as
overlap
f
actor
(
C
1
)+
coverage
f
actor
(
C
1
)+
usage
f
actor
(
C
1
)=3(=(0
+0+0)+1+2).Thefitnessvalueof
C
1
is thus (0.266 + 0.31)/3 (=
0.192). The fitness values of all the chromosomes in the four populations are
calculated with their results shown in Table 5.
STEP 4: The crossover operator is executed on the populations. Take
C
1
and
C
5
in
Population
3
as an example. Assume for the control genes, the one-
point crossover operator selects the first number as the crossover point and for
the parametric genes,
d
is set at 0.35.The following four candidate offspring
chromosomes are generated:
C
1
: 1 1 1 1, 0 3 5, 3 5 10, 6 13 16, 15 20 20
C
5
: 1 1 1 1, 0 3 8, 5 8 12, 10 12 16, 12 16 16
