Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Query Data With Fuzzy Information In Object-Oriented Databases An Approach The Semantic Neighborhood Of Hedge Algebras

Query Data With Fuzzy Information In Object-Oriented Databases An Approach The Semantic Neighborhood Of Hedge Algebras

Ratings: (0)|Views: 46 |Likes:
Published by ijcsis
In this paper, we present an approach for handling attribute values of object classes with fuzzy information and uncertainty in object-oriented database based on theory hedge algebraic. In this approach, semantics be quantified by quantitative semantic mapping of hedge algebraic that still preserving in order semantics may allow manipulation data on the real domain of attribute in relation with the semantics of linguistic. And then, evaluating semantics, searching information uncertainty, fuzziness and classical data entirely consistent based on the ensuring homogeneity of data types. Hence, we present algorithm that allow the data matching helping the requirements of the query data.
In this paper, we present an approach for handling attribute values of object classes with fuzzy information and uncertainty in object-oriented database based on theory hedge algebraic. In this approach, semantics be quantified by quantitative semantic mapping of hedge algebraic that still preserving in order semantics may allow manipulation data on the real domain of attribute in relation with the semantics of linguistic. And then, evaluating semantics, searching information uncertainty, fuzziness and classical data entirely consistent based on the ensuring homogeneity of data types. Hence, we present algorithm that allow the data matching helping the requirements of the query data.

More info:

Published by: ijcsis on Jun 05, 2011
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

06/05/2011

pdf

text

original

 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 5, May 2011
Query Data With Fuzzy Information In Object-Oriented Databases An Approach The Semantic Neighborhood Of Hedge Algebras
 
Doan Van Thang
Korea-VietNam Friendship Information Technology CollegeDepartment of Information systems, Faculty of Computer Science
Da Nang City, Viet Nam
vanthangdn@gmail.com
 
Doan Van Ban
Institute of Information Technology, Academy Science andTechnology of Viet Nam.Ha Noi City, Viet Nam
 
 Abstract -
In this paper, we present an approach for handlingattribute values of object classes with fuzzy information anduncertainty in object-oriented database based on theoryhedge algebraic. In this approach, semantics be quantified byquantitative semantic mapping of hedge algebraic that stillpreserving in order semantics may allow manipulation dataon the real domain of attribute in relation with the semanticsof linguistic. And then, evaluating semantics, searchinginformation uncertainty, fuzziness and classical data entirelyconsistent based on the ensuring homogeneity of data types.Hence, w
 
e present algorithm that allow the data matchinghelping the requirements of the query data.
 
I.
 
I
 NTRODUCTION
 In approach interval value [2], we consider toattributive values object class is interval values and theinterval values are converted into sub interval in [0, 1]respectively and then we perform matching interval this.However, attributive value of the object in the fuzzyobject-oriented database is complex: linguistic values,reference to objects (this object may be fuzzy),collections,… Thus matching data also become morecomplex. Hence, query information method proposed in[2] is not satisfy requirements for the case of this data yet.In this paper, we research has expanded for handlingattribute value is linguistic value. There are manyapproaches on handling fuzzy information with linguisticsematic that researchers interests [1], [3]. We based onapproach hedge algebra, where linguistic semantic isobtained by considering the terms as expressed by the partof order relation. In this approach linguistic value is datawhich is not label of fuzzy set representation sematic of linguistic value. Using quantitative semantics mapping of hedge algebra to transfer linguistic values into real valuesthat preserve in order semantics may allow manipulationdata on the real domain of attribute in relation with thesemantics of linguistic.The paper is organized as follows: Section 2 presenting the basic concepts relevant to hedge algebraicas the basis for the next sections; section 3 proposing
SASN
(
S
earch
A
ttributes in the
S
emantic
N
eighborhood)and
SMSN
(
S
earch
M
ethod in the
S
emantic
N
eighborhood) algorithms for searching data fuzzyconditions for both attributes and methods; section 4 presenting examples for searching data with fuzzyinformation, and finally conclusion.II.
 
F
UNDAMENTAL CONCEPTS
 In this section, we present some fundamentalconcepts related to hedge algebra [5].Let hedge algebra
 X = (
 X 
 , G, H,
≤  
)
, where
X
=
 LDom(X)
,
G
= {
1
, c
-
,
W
, c
+
,
0
} is set generator terms,
 H 
isa set of hedge considered as a one-argument operationsand
relation on terms (fuzzy concepts) is a relation order “induced” from natural semantics on
 X 
. Set
 X 
is generatedfrom
G
by means of one-argument operations in
 H 
. Thus,a term of 
 X 
represented as
 x = h
n
h
n-1
.......h
1
 x
,
 x
 
G
. Set of terms is generated from the an
 X 
term denoted by H
(x)
.Let set hedges
 H = H 
− 
 
∪ 
+
 ,
where
 H 
+
= {h
1
 ,..., h
 p
 } and  H 
-
= {h
-1
 , ..., h
-q
 }
are linearly ordered
 ,
with
h
1
< .. .< h
 p
 and h
-1
< .. .< h
-q
 ,
where
p, q >1
, we have the followingdefinitions related:
Definition 2.1
An
 fm
: X
[0,1] is said to be a fuzzinessmeasure of terms in X if:(1) fm is called complete, that is
u
X,
,0
()()
iqipi
 fmhufmu
− ≤
=
 (2) if x is precise, that is H(x) = {x} then
 fm(x) = 0
.Hence
 fm(0)=fm(W)=fm(1)=0
.(3)
x,y
X,
h
H,()()()()
 fmhxfmhy fmxfmy
=
,
This proportion is called the fuzziness measure of the hedge hand denoted by
μ 
(
h
)
.
Definition 2.2
(Quantitative semantics function
ν 
)
 
Let
 fm
is fuzziness measure of 
 X 
, quantitativesemantics function
v
on
 X 
is defined as follows:(1)
v(
)=
θ 
= fm(c
-
),
ν 
(c
− 
) =
θ 
-
α 
.fm(c
-
) and 
ν 
(c
+
) =
θ 
+
α 
.fm(c
+
)
(2) If 1
j
p then:
1
()()()()()()
 j jjijji
vhxvxSignhxfmhxhxfmhx
ω 
=
= + ×
 (3) If -q
j
-1 then:
1
()()()()()()
 jjijjij
vhxvxSignhxfmhxhxfmhx
ω 
=
= + ×
 
HU
http://sites.google.com/site/ijcsis
U
 ISSN 1947-5500
37http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 5, May 2011
Where
:
{ }
1()1()()(),2
 jjqj
hxSignhxSignhhx
ω β α α β 
= +
 
Definition 2.3
Invoke
 fm
is fuzziness measure of hedgealgebra X,
 f 
: X -> [0, 1].
x
X, denoted by
 I(x)
 
[0, 1]and |I(x)| is measure length of I(x).A family J = {I(x):x
X} called the partition of [0, 1]if:(1): {I(c
+
), I(c
-
)} is partition of [0, 1] so that|I(c)| =
 fm
(c), where c
{c
+
, c
-
}.(2): If I(x) defined and |I(x)| = fm(x) then{I(hix): I = 1...p+q}is defined as a partition of I(x) so thatsatisfy conditions: |I(h
i
x)| =
 fm
(h
i
x) and |I(h
i
x)| is linear ordering.Set {I(h
i
x)} called the partition associated with theterms x. We have
1
()()()
 pqii
 IhxIxfmx
+=
= =
 
Definition 2.4
Set X
=
{ }
:
 xXx
=
, consider P
=
{ }
():
 Ixx
is a partition of [0, 1]. Its said that u equalv at k level, denoted by u =
v, if and only if 
 I(u)
and
 I(v)
 together included in fuzzy interval k level. Denote
u, v
X, :()
kk
uvPIu
= Δ Δ
and ()
 Iv
⊆ Δ
.III.
 
D
ATA SEARCH METHOD
 Let fuzzy class C = ({a
1
, a
2
, …, a
n
}, {M
1
, M
2
, …,M
m
});
o
is object of fuzzy class C. Denoted o.a
i
isattribute value of 
o
on attribute a
i
(1
in
) and o.M
 j
isvalue method of 
o
(
1
 jm
).In [2] we presented the attribute values are 4 cases: precise value; imprecise value (or fuzzy); object;collection. In this paper, we only interested in handingcase 1 and 2: precise value and imprecise value (fuzzyvalue) and to see precise value is particular case of fuzzyvalue. Fuzzy value is complex and linguistic label is oftenused to represent the value of this type. Domain fuzzyattribute value is the union two components:
 Dom(a
i
)
=
CDom(a
i
)
 
 
FDom(a
i
)
(
1
in
).Where:-
 
CDom(a
i
)
: domain crisp values of attributea
i
.-
 
FDom(a
i
)
: domain fuzzy values of attributea
i
.
 A.
 
 Neighborhood level k 
We can get fuzzy interval of terms length k as thesimilarity between terms. It means that the term thatrepresentative value of them depending on fuzzy interval
level k 
is similar 
level k 
. However, to build the fuzzyinterval
level k 
, representative value of terms
 x
havelength less than k is always in the end of fuzzy interval
level k 
. Hence, when determining
neighborhood level k,
 we expect representative value it must be inner point of 
neighborhood level k 
.Based on fuzzy interval level
and
k+1
we constructa partition of the domain [0, 1] following as [8]:(1)
Similar level 1
: with
= 1, fuzzy interval
level 1
 including
 I 
(
c
− 
) and
 I 
(
c
+
). fuzzy interval
level 2
on interval
 I 
(
c
+
) is
 I 
(
h
-q
c
+
)
 
 I 
(
h
-q+
1
c
+
) ...
 
 I 
(
h
-2
c
+
)
 
 I 
(
h
-1
c
+
)
 
υ 
 A
(
c
+
)
 
 I 
(
h
1
c
+
)
 
 I 
(
h
2
c
+
)
...
 
 I 
(
h
 p-
1
c
+
)
 
 I 
(
h
 p
c
+
). Meanwhile,we construct partition at
similar level 1
include theequivalence classes following:
S
(
0
) =
 I 
(
h
 p
c
− 
);
S
(
c
− 
)=
 I 
(
c
− 
) \ [
 I 
(
h
-q
c
− 
)
 
 I 
(
h
 p
c
− 
)];
S
(
) =
 I 
(
h
-q
c
− 
)
 
 I 
(
h
-q
c
+
);
S
(
c
+
) =
 I 
(
c
+
) \ [
 I 
(
h
-q
c
+
)
 
 I 
(
h
 p
c
+
)] and
S
(
1
) =
 I 
(
h
 p
c
+
).We see that except the two end points
υ 
 A
(
0
) = 0 and
υ 
 A
(
1
) = 1, representative values
υ 
 A
(
c
− 
),
υ 
 A
(
) and
υ 
 A
(
c
+
)are inner point corresponding of classes
similar level 1
 
S
(
c
− 
),
S
(
) and
S
(
c
+
).(2)
Similar level 2
: with
= 2, fuzzy interval
level 2
 including
 I 
(
h
i
c
+
) and
 I 
(
h
i
c
-
) with -q
 
i
p. We haveequivalence classes following:
S
(
0
) =
 I 
(
h
 p
h
 p
c
− 
);
S
(
h
i
c
− 
) =
 I 
(
h
i
c
− 
) \ [
 I 
(
h
-q
h
i
c
− 
)
 
 I 
(
h
 p
h
i
c
− 
)];
S
(
) =
 I 
(
h
-q
h
-q
c
− 
)
 
 I 
(
h
-q
h
-q
c
+
);
S
(
h
i
c
+
) =
 I 
(
h
i
c
+
) \ [
 I 
(
h
-q
h
i
c
+
)
 
 I 
(
h
 p
h
i
c
+
)] and
S
(
1
) =
 I 
(
h
 p
h
 p
c
+
), with -q
 
i
p.By the same, we can construct partition equivalenceclasses
level k 
at any. However, in fact,
 
4 and it meansthat there is maximum 4 hedges consecutive action onto primary terms
 
c
and c
+
. Precise and fuzzy values will beat the
similar level k 
if the representative value of their inthe same class
similar level k 
.Hence,
neighborhood level k of 
uzzy concept 
isdetermining following: Assuming partition the class
similar level k 
is intervals
S
(
 x
1
),
S
(
 x
2
), …,
S
(
 x
m
).Meanwhile, every fuzzy value
 fu
is only and only belongto a similar class. Instance for 
S
(
 x
i
) and called
neighborhood level k 
of 
 fu
and denoted by ()
FRNfu
.
 B.
 
 Relation matching on domain of fuzzy attribute value
Based on the concept neighborhood, we give thedefinition of the relation matching between terms in thedomain of the fuzzy attribute value.
Definition 3.1
Let fuzzy class C determine on the set of attributes Aand methods M, a
i
A. o
1
, o
2
 
C. We say that
12
..
iki
oaoa
=
and equal
level k 
if:(1)
 
If 
12
.,.()
iii
oaoaCDoma
then
12
..
ii
oaoa
=
or existence
()
FRNx
such that
12
.,.()
ii
oaoaFRNx
.(2)
 
If 
1
.
i
oa
or 
2
.
i
oa
()
i
FDoma
, instance for 
1
.
i
oa
 then we have to
21
.(.)
iki
oaFRNoa
.(3)
 
If 
1
.
i
oa
,
2
.
i
oa
 
()
i
FDoma
then
21
(.)(.)
kiki
FRNoaFRNoa
=
.
Definition 3.2
Let fuzzy class C determine on the set of attributes Aand methods M, a
i
A. o
1
, o
2
 
C. We say that
12
..
iki
oaoa
if:(1)
 
If 
12
.,.()
iii
oaoaCDoma
then
12
..
ii
oaoa
.(2)
 
If 
1
.
i
oa
and
2
.
i
oa
 
()
i
FDoma
then we have to
38http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 5, May 2011
12
.(.)
iki
oaFRNoa
.(3)
 
If 
1
.
i
oa
,
2
.
i
oa
()
i
FDoma
then
12
(.)(.)
kiki
FRNoaFRNoa
 
C.
 
 Algorithm search data approach to semanticneighborhood 
In [2] we presented the structure of fuzzy OQLqueries are considered as:
select 
<attributes>/<methods>
 from
<class>
where
<
 fc
>, where <
 fc
> are fuzzy conditionsor combination of fuzzy condition that allow using of disjunction or conjunction operations.In this paper, we use approaching to semanticneighborhood for determinating the truth value of the <
 fc
>and associated truth values.Example, we consider query following “
show allstudents are possibly young age
”. To answer this query,we perform following:+
Step 1
: We construct intervals
similar level k 
,
 
4
 
 because it’s a maximum 4 hedges consecutive action onto primary terms
 
c
and c
+
.+
Step 2
: Determine
neighborhood level k 
of fuzzycondition. In the above query, fuzzy condition is
possibly young
should
neighborhood level 2
of 
possibly young
is
FRN 
2
(
possibly young
)
, anddetermine
neighborhood level 2
of fuzzy attribute value is
FRNA
2
(attr)
. At last based on definition 3.1, we performdata matching two
neighborhood level 2
of 
 
FRNA
2
(attr)
and
FRN 
2
(
 possibly young 
)
.Without loss of generality, we consider on casesmultiple fuzzy conditions with notation follow as:-
ϑ 
is
 AND or 
 
OR
operation.-
i
 fzvalue
is fuzzy values of the i attribute.On that basis, we built the
SASN
algorithms
SASN algorithm:
 
search data in cases multiple fuzzyconditions for attribute with operation
ϑ 
.
Input
: A class C = ({a
1
, a
2
, …, a
n
}, {M
1
, M
2
, …, M
m
}),C = { o
1
, o
2
,…, o
n
}.where a
i
, i = 1…p is attribute, M
 j
is methods.
Output
: Set of objects o
C satisfy condition
1
 pi
ϑ 
=
(
o.a
i
=
i
 fzvalue
).
Method
// Initialization.(1)
For
i = 1
to
p
do
 
(2)
Begin
 (3) Set
i
a
G
= {
0
,
i
a
c
,
,
+
i
a
c
,
1
};
 
i
a
 H 
=
+
i
a
 H 
i
a
 H 
. Where
+
i
a
 H 
={
h
1
,
h
2
},
i
a
 H 
={
h
3
,
h
4
}, with
h
1
<
h
2
and
h
3
>
h
4
. Select the fuzzymeasure for the generating term and hedge.(4)
]max,[min
iii
aaa
 D
=
//
i
a
min
,
i
a
max
: min andmax value of domain a
i
.(5) ()()
iiiii
aaaaa
FDHcHc
+
=
.(6)
End
 
(7) Determine
intervals level
of fuzzy condition:
kQ 
.
 
// Partition
i
a
 D
into interval
similar level k 
.(8)
k
=
kQ 
;
//
level partition largest with k 
= 4
 
(9)
For
i = 1
to
p
do
 (10)
For
 
j = 1
to
2
5
(k-1)
do
 (11) Construct intervals similar level
:
()
i
aj
Sx
;
 // Determine
neighborhood level k 
of .
i
oa
.
 (12)
For
each o
C
do
 (13)
For
i = 1
to
p
do
(14)
Begin
 
(15)
t=0;
 (16)
Repeat
 (17)
t=t+1;
 (18)
Until
 
.
i
oa
 
 
()
i
a
Sx
or
t >
2
5
(k - 1)
;
 (19)
()
ii
FRNAatt
=
()
ii
FRNAatt
()
i
a
Sx
;
 (20)
End
// Determine
neighborhood level k 
of 
i
 fzvalue
.
 
(21)
For
i = 1
to
p
do
 (22)
Begin
 
(23)
t=0;
 (24)
Repeat
 (25)
t=t+1
;(26)
Until
 
i
 fzvalue
 
 
()
i
a
Sx
or
t > 2
5
(k - 1)
;
(27)()
ii
FRNfzvalue
=
()
ii
FRNfzvalue
()
i
a
Sx
;
 (28)
End
 (29)
result=
;
 (30)
For
each o
C
do
 (31)
if
1
 pi
ϑ 
=
(
()
i
FRNAattr 
()
i
FRNAfzvalue
) 
then
result=result
{o};
(32)
Return result;
Similar to the method we have
SMSN
algorithmfollowing:
SMSN algorithm:
 
search data cases single fuzzyconditions for method.
Search data in this case, the first we determine
neighborhood level k 
fuzzy conditions of method is
FRNP
(fzpvalue)
. Further, we determine
neighborhood level k 
of attributes which method handing:
FRNA
(attr1),FRNA
(attr2), …, FRNA
(attrn)
. We choose the functioncombination of hedge algebras being consistent withmethod that it operate. Then,
neighborhood level k 
of function combination is
FRNPA
(x)
.At last based on definition 3.1, we perform datamatching two
neighborhood level k 
of 
FRNP
(fzpvalue)
and
FRNPA
(x).
 
Input
: A class C = ({a
1
, a
2
, …, a
n
}, {M
1
, M
2
, …, M
m
}),C = { o
1
, o
2
,…, o
n
}.where a
i
, i = 1…p is attribute, M
 j
is methods.
Output
: Set of objects o
C satisfy condition(
o.M 
i
= fzpvalue
).
Method
39http://sites.google.com/site/ijcsis/ISSN 1947-5500

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->