(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 1, January 2011
Similarly the
α
cut of the fuzzy interval [
t
1
a, t
1
, t
2
, t
2
+a
] isa closed interval [
t
1
+
(
α

1)
.a, t
2
+
(1

α
)
.a
].The core of a fuzzy number
A
is the set of elements of
A
having membership value one i.e.Core(
A
) = {(
x
,
A
(
x
);
A
(
x
) = 1}For every fuzzy set
A
,
A
=
U
]1,0[
∈
α α
A
where
α
A
(
x
) =
α
.
α
A
(
x
), and
α
A
is a special fuzzy set [4]For any two fuzzy sets
A
and
B
and for all
α
∈
[0, 1],i)
α
(
A
∪
B
) =
α
A
∪
α
B
ii)
α
(
A
∩
B
) =
α
A
∩
α
B
For any two fuzzy numbers
A
and
B
, we say themembership functions
A
(
x
) and
B
(
x
) are similar to each other if the slope of the left reference function of
A
(
x
) is equal to thethat of
B
(
x
) and the slope of right reference of
A
(
x
) is equal thatof
B
(
x
). Obviously for any two fuzzy numbers
A
and
B
havingsimilar membership functions
⏐
α
A
⏐
=
⏐
α
B
⏐
,
∀α∈
[0, 1]
B. Some Definition related to Fuzzy Locally Frequent set
Let T = <
t
o
, t
1
,…………
> be a sequence of imprecise or fuzzytime stamps over which a linear ordering < is defined where
t
i
<
t
j
means
t
i
denotes the core of a fuzzy time which is earlierthan the core of another fuzzy time stamp
t
j.
For the sake of convenience, we assume that all the fuzzy time stamps arehaving similar membership functions. Let
I
denote a finite setof items and the transaction database
D
is a collection of transactions where each transaction has a part which is a subsetof the itemset
I
and the other part is a fuzzy timestampindicating the approximate time in which the transaction hadtaken place. We assume that
D
is ordered in the ascendingorder of the core of fuzzy time stamps. For fuzzy time intervalswe always consider a fuzzy closed intervals of the form [
t
1
a t
1
,
t
2
, t
2
+a
] for some real number
a
. We say that a transaction is inthe fuzzy time interval [
t
1
a, t
1
,
t
2
, t
2
+a
] if the
α
cut of the fuzzytime stamp of the transaction is contained in
α
cut of [
t
1
a, t
1
,
t
2
, t
2
+a
] for some user’s specified value of
α
.We define the local support of an itemset in a fuzzy timeinterval [
t
1
a, t
1
,
t
2
, t
2
+a
] as the ratio of the number of transactions in the time interval [
t
1
+
(
α

1)
.a, t
2
+
(1

α
)
.a
]containing the itemset to the total number of transactions in[
t
1
+
(
α

1)
.a, t
2
+
(1

α
)
.a
] for the whole data base
D
for a givenvalue of
α
. We use the notation (
X
) todenote the support of the itemset
X
in the fuzzy time interval[
t
],,,[
2211
at t t at
Sup
+−
1
a, t
1
,
t
2
, t
2
+a
]. Given a threshold
σ
we say that an itemset
X
is frequent in the fuzzy time interval [
t
1
a, t
1
,
t
2
, t
2
+a
] if (
X
)
≥
(
σ
/100)* tc where tc denotes the totalnumber of transactions in
D
that are in the fuzzy time interval[
t
],,,[
2211
at t t at
Sup
+−
1
a, t
1
,
t
2
, t
2
+a
]. We say that an association rule
X
⇒
Y
, where
X
and
Y
are item sets holds in the time interval [
t
1
a, t
1
,
t
2
, t
2
+a
]if and only if given threshold
τ
,
0.100 / )( / )(
],,,[],,,[
22112211
τ
≥∪
+−+−
X SupY X Sup
at t t at at t t at
and
X
∪
Y
is frequent in [
t
1
a, t
1
,
t
2
, t
2
+a
]. In this case we saythat the confidence of the rule is
τ
.For each locally frequent item set we keep a list of fuzzytime intervals in which the set is frequent where each fuzzyinterval is represented as [
starta, start, end, end+a
] where
start
gives the approximate starting time of the time intervaland
end
gives the approximate ending time of the timeinterval.
end
–
start
gives the length of the core of the fuzzy timeinterval. For a given value of
α
of two intervals [
start
1
a,
start
1
,
end
1
,
end
1
+a] and [
start
2
a,
start
2
,
end
2
,
end
2
+a] are nonoverlapping if their
α
cuts are nonoverlapping.IV.
PROPOSED ALGORITHM
A. Generating Fuzzy Locally Frequent Sets
While constructing locally frequent sets, with each locallyfrequent set a list of fuzzy timeintervals is maintained in whichthe set is frequent. Two user’s specified thresholds
α
and
minthd
are used for this. During the execution of the algorithmwhile making a pass through the database, if for a particularitemset the
α
cut of its current fuzzy timestamp, [
α
Lcurrent,
α
Rcurrent
] and the
α
cut, [
α
Llastseen,
α
Rlastseen
] of its
fuzzytime, when it was last seen overlap then the current transactionis included in the current timeinterval under considerationwhich is extended with replacement of
α
Rlastseen
by
α
Rcurrent
; otherwise a new timeinterval is started with
α
Lcurrent
as the starting point. The support count of the itemset in the previous time interval is checked to see whether it isfrequent in that interval or not and if it is so then it is fuzzifiedand added to the list maintained for that set. Also for the fuzzzylocally frequent sets over fuzzy time intervals, a minimum corelength of the fuzzy period is given by the user as
minthd
andfuzzy time intervals of core length greater than or equal to thisvalue are only kept. If
minthd
is not used than an itemappearing once in the whole database will also become locallyfrequent a over fuzzy point of time.Procedure to compute
L
1
, the set of all fuzzy locally frequentitem sets of size 1.For each item while going through the database we alwayskeeps an
α
cut
α
lastseen
which is [
α
Llastseen,
α
Rlastseen
] thatcorresponds to the fuzzy time stamp when the item was lastseen. When an item is found in a transaction and the fuzzytimestamp is
tm
and if its
α
cut
α
tm
=[
α
Ltm,
α
Rtm
] has emptyintersection with [
α
Llastseen,
α
Rlastseen
], then a new timeinterval is started by setting
start
of the new time interval as
α
Ltm
and
end
of the previous time interval as
α
Rlastseen
. Theprevious time interval is fuzzified provided the support of theitem is greater than
minsup
. The fuzzified interval is thenadded to the list maintained for that item provided that theduration of the core is greater than
minthd
. Otherwise
α
Rlastseen
is set to
α
Rtm
, the counters maintained for counting
141http://sites.google.com/site/ijcsis/ISSN 19475500