IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Hf2412861289
1. P.M.Sitharselvam , T.Priya, T.Ramachandran/ International Journal of Engineering Research
and Applications (IJERA) ISSN: 2248-9622 www.ijera.com
Vol. 2, Issue4, July-August 2012, pp.1286-1289
Anti Q-Fuzzy KU - Ideals In KU- Algebras And Its Lower Level
Cuts
1
P.M.Sitharselvam , 2 T.Priya, 3 T.Ramachandran
1,2
Department of Mathematics, PSNA College of Engineering and Technology,
Dindigul-624 622, Tamilnadu , India.
3
Department of Mathematics, Government Arts College(Autonomous),
Karur– 639 005,Tamilnadu , India
Abstract
In this paper, we introduce the concept of 2. 0 * x = x
Anti Q-fuzzy KU-ideals of KU-algebras, lower level 3. x * 0 = 0
cuts of a fuzzy set and prove some results . We show 4. x * y = 0 = y * x implies x = y , for all x, y , z X .
that a Q-fuzzy set of a KU-algebra is a KU-ideal if In X we can define a binary operation by
and only if the complement of this Q-fuzzy set is an x y if and only if y * x = 0. Then (X,*,0) is a
anti Q-fuzzy KU-ideal. Also we discussed few results KU-algebra if and only if it satisfies that
of Anti Q-fuzzy subalgebras of KU-algebra. i. (y * z) * (x * z) ≤ (x * y)
ii. 0 ≤ x
Keywords:- KU-algebra,fuzzy KU- ideal,Anti Q- iii. x ≤ y , y ≤ x implies x = y.
fuzzy KU-ideal, Q-fuzzy subalgebra, Anti Q- fuzzy iv. x y if and only if y * x = 0, for all x, y , z X.
subalgebra, lower level cuts. In a KU-algebra,the following identities are true[14]:
AMS Subject Classification (2000): 20N25, 03E72, 1. z * z = 0.
03F055 , 06F35, 03G25. 2. z * (x * z) = 0
3. x y y * z ≤ x * z
1.Introduction 4. z * ( y * x ) = y * (z * x)
Y.Imai and K.Iseki introduced two classes of 5. y * [ (y * x) * x ] = 0 , for all x, y , z X.
abstract algebras : BCK-algebras and BCI – Example 2.1
algebras[6,7]. It is known that the class of BCK- Let X = { 0 , a , b, c, d } be a set with a binary
algebras is a proper subclass of the class of BCI- operation * defined by the following table
algebras. Q.P.Hu and X .Li introduced a wide class of
abstract BCH-algebras[4,5]. They have shown that it * 0 a b c d
is the class of BCI-algebras. J.Neggers, S.S.Ahn and 0 0 a b c d
H.S.Kim introduced Q-algebras which is a 0 0 b c d
generalization of BCK / BCI algebras and obtained b 0 a 0 c c
several results[10].W.A.Dudek and Y.B.Jun studied c 0 0 b 0 b
fuzzy ideals and several fuzzy structures in BCC- d 0 0 0 0 0
algebras are considered[2,3]. C.Prabpayak and
U.Leerawat introduced a new algebraic structure
Then Clearly ( X , * , 0 ) is a KU-algebra.
which is called KU-algebras and investigated some
properties[11].Samy M.Mostafa and Mokthar A.
Definition 2.2 [11]
Abdel Naby introduced fuzzy KU-ideals in KU-
Let ( X , * , 0) be a KU-algebra. A non empty subset
algebras[14].R.Biswas introduced the concept of Anti
I of X is called KU- ideal of X if it satisfies the
fuzzy subgroups of groups[1]. Modifying his idea, in
following conditions
this paper we apply the idea of KU-algebras . We
(1) 0I
introduce the notion of Anti Q-fuzzy KU-ideals of
KU-algebras and Anti Q-fuzzy subalgebras of KU- (2) x * (y * z) I and y I x * z I
algebras. for all x , y, z X .
Remark: From Example 2.1, It is clear that I1 = {0,a}
2. Preliminaries
and I2 = {0,a,b} are KU-ideals of X.
In this section we site the fundamental
Definition 2.3 [16]
definitions that will be used in the sequel.
Definition 2.1[11] Let X be a non-empty set. A fuzzy subset
A non empty set X with a constant 0 and a of the set X is a mapping : X [0, 1].
binary operation * is called a KU-algebra if it Definition 2.4 [8,9]
satisfies the following axioms. Let Q and G be any two sets.A mapping
1. (x * y) *[(y * z) * (x * z) ] = 0 : G x Q [0, 1] is called a Q –fuzzy set in G.
1286 | P a g e
2. P.M.Sitharselvam , T.Priya, T.Ramachandran/ International Journal of Engineering Research
and Applications (IJERA) ISSN: 2248-9622 www.ijera.com
Vol. 2, Issue4, July-August 2012, pp.1286-1289
(ii) µc ( x * z, q) = 1 - µ(x * z, q)
Definition 2.5 ≤ 1 – min { µ ( x * ( y * z) , q) , µ ( y ,q) }
A Q- fuzzy set in X is called a Q-fuzzy = 1 – min{1- µc ( x * (y * z) , q) ,1 - µc(y,q) }
KU- ideal of X if = max { µc ( x * ( y * z) , q) , µc ( y , q) }
(i) (0,q) ≥ (x,q) That is,
(ii) (x* z ,q) ≥ min{(x *( y * z),q), (y,q)}, µc ( x * z , q) ≤ max { µc ( x* ( y * z) , q) , µc ( y , q) }.
for all x,y, z X and q ∈ Q. Thus µc is an anti Q-fuzzy KU-ideal of X. The
converse also can be proved similarly.
Definition 2.6 [9]
If µ be a Q-fuzzy set in set X then the Theorem 3.3
complement denoted by µc is the Q-fuzzy subset of X Let µ be an anti Q-fuzzy KU-ideal of
given by µc (x,q) = 1-µ(x,q), for all x,y X and q ∈ Q. KU – algebra X. If the inequality x * y ≤ z , then
µ(y,q) ≤max{µ(x,q),µ(y,q)} for all x,y,zX and q∈Q.
3.ANTI Q-FUZZY KU-IDEALS
Definition 3.1 Proof:
Assume that the inequality x * y ≤ z for all
A Q-fuzzy set of a KU-algebra X is called
an anti Q-fuzzy KU-ideal of X, if x,y,z X and q ∈ Q. Then z * ( x * y) = 0 .
Now, µ ( y , q ) = µ ( 0 * y , q )
(i) (0,q) ≤ (x,q)
≤ max { µ ( 0 * (x * y) , q ), µ (x,q) }
(ii) (x *z,q) ≤ max { ((x * (y * z)) , q ), ( y , q )},
= max { µ (x * y,q ), µ (x,q)}
for all x,y,z X and q ∈ Q.
≤ max{max{ (x * (z * y) , q ), ( z , q )} , µ(x,q)}
Example 3.1
Let X = { 0 , 1 , 2, 3, 4 } be a set with a binary = max{max{ (z * (x * y) , q ), ( z , q )} , µ(x,q)}
operation * defined by the following table = max { max { ( 0, q ), ( z , q )} , µ(x,q)}
* 0 1 2 3 4 = max { ( z , q )} , µ(x,q)}
0 0 1 2 3 4 µ (y,q) ≤ max {µ(x,q),µ(y,q)}.
1 0 0 2 2 4
2 0 0 0 1 4 Theorem 3.4
If µ is an anti Q-fuzzy KU-ideal of KU–
3 0 0 0 0 4
4 0 0 0 1 0 algebra X, then for all x, y X and q ∈ Q,
µ ( x * (x * y), q ) ≤ µ ( y , q )
Then Clearly ( X , * , 0 ) is a KU-algebra.
Let t0, t1, t2 ∈ [ 0, 1] be such that t0 < t1 < t2 .Define a Q- Proof:
fuzzy set µ : X x Q [0, 1] by µ(0,q) = t0, µ(1,q) = t1 = Let x,y ∈ X and q ∈ Q.
µ(2,q), µ(3,q) = t2 = µ(4,q) ,by routine calculations µ is µ (x*(x*y),q) ≤ max{ µ(x* (y*( x * y)), q ), µ( y, q) }
an anti Q-fuzzy KU- ideal of X where q∈Q. = max { µ ( x * (x * ( y * y)), q ), µ( y, q) }
= max { µ ( x * (x * 0), q ), µ( y, q) }
Theorem 3.1 = max { µ ( x * 0, q ), µ( y, q) }
Every Anti Q-fuzzy KU- ideal of a KU- = max { µ ( 0, q ), µ( y, q) }
algebra X is order preserving. = µ (y,q)
µ ( x * (x * y), q ) ≤ µ ( y , q )
Proof
Let be an anti Q-fuzzy KU- ideal of a KU- 4.LOWER LEVEL CUTS IN ANTI Q-
algebra X and let x , y X and q ∈ Q. FUZZY KU-IDEALS OF KU-ALGEBRA
Then (x,q) = (0 * x , q)
≤ max { (0 * (y * x) , q) , (y,q)} Definition 4.1 [12]
= max { (0 * 0 , q) , (y,q) } Let be a Q-fuzzy set of X. For a fixed t
= (y,q) [0, 1], the set t ={x X (x,q) ≤ t for all q ∈ Q} is
(x,q) ≤ (y,q). called the lower level subset of .
Clearly t t =X for t[0,1] if t1 < t2 , then t1 t2.
Theorem 3.2
is a Q-fuzzy KU-ideal of a KU-algebra X if Theorem 4.1
and only if µc is an anti Q-fuzzy KU-ideal of X. If µ is an anti Q-fuzzy KU-ideal of KU-
Proof : algebra X,then µt is an KU-ideal of X for every
Let µ be a Q-fuzzy KU- ideal of X and let x , y , z t [0,1] .
X and q ∈ Q. Proof
(i) c
µ (0,q) = 1 - µ(0,q) Let µ be an anti Q-fuzzy KU-ideal of KU-
≤ 1 - µ ( x,q) algebra X.
= µc ( x,q) (i) Let x ∈ µt µ( x, q ) ≤ t.
That is µc(0,q) ≤ µc ( x,q) . µ(0,q) = µ ( y * 0, q )
1287 | P a g e
3. P.M.Sitharselvam , T.Priya, T.Ramachandran/ International Journal of Engineering Research
and Applications (IJERA) ISSN: 2248-9622 www.ijera.com
Vol. 2, Issue4, July-August 2012, pp.1286-1289
≤ max { µ ( y * (x * 0),q ) , µ(x,q) } µ (x,q) ≤ t .
= max { µ ( y * 0),q ) , µ(x,q) } x ∈ µt.
= max { µ (0),q ) , µ(x,q) } Therefore A(a,b) ⊆ µt .
= µ(x,q) ≤ t. Conversely suppose that A(a,b) ⊆ µt .
0 ∈ µt . Obviously 0 ∈ A(a,b) ⊆ µt for all a,b ∈ X.
( ii) Let x * (y * z) ∈ µt and y ∈ µt , for all x,y,z X Let x,y,z ∈ X be such that x * (y * z) ∈ µt and y ∈ µt .
and q ∈ Q. Since (x * ( y * z )) * (y * ( x * z )) = (x * ( y * z )) *
µ (x * (y * z) ,q ) ≤ t and µ ( y,q ) ≤ t. ( x * ( y * z )) = 0.
µ ( ( x * z),q ) ≤ max { ((x * (y * z)) , q ), ( y , q )} We have x * z ∈ A ( x * ( y * z ) , y ) ⊆ µt .
≤ max {t,t} = t. µt is an KU- ideal of X.
x * z ∈ µt. Hence,by theorem 4.2, µ is an anti Q-fuzzy KU-ideal
Hence µt is an KU- ideal of X for every t [0,1]. of X.
Theorem 4.2 Theorem 4.4
Let µ be a Q-fuzzy set of KU- algebra X.If Let µ be a Q-fuzzy set in KU-algebra X.If µ
for each t ∈ [0,1], the lower level cut µt is an KU- is an anti Q-fuzzy KU-ideal of X then
ideal of X, then µ is an anti Q- fuzzy KU-ideal of X. ∀ 𝑡 ∈ 0,1 𝜇𝑡 ≠ ∅ 𝜇𝑡 = 𝐴 𝑎, 𝑏 .
𝑎,𝑏 ∈ 𝜇 𝑡
Proof
Let µt be an KU-ideal of X. Proof :
If µ(0,q) > µ(x,q) for some x ∈ X and q ∈ Q.Then Let t ∈ [0,1] be such that 𝜇 𝑡 ≠ ∅. Since 0 ∈ µt , we
1
µ(0,q) > t0 > µ(x,q) by taking t0 = { µ(0,q) + µ(x,q) }.
2 have 𝜇 𝑡 ⊆ 𝑎 ∈ 𝜇 𝑡 𝐴 𝑎, 0 ⊆ 𝑎,𝑏 ∈ 𝜇 𝑡 𝐴 𝑎, 𝑏 .
Hence 0 ∉ µt0 and x ∈ µt0 , which is a contradiction. Now, let x ∈ 𝑎,𝑏 ∈ 𝜇 𝑡 𝐴 𝑎, 𝑏 .
Therefore, µ(0,q) ≤ µ(x,q). Then there exists (u,v) ∈ µt such that x ∈ A (u,v) ⊆ µt
Let x,y,z ∈ X and q ∈ Q be by theorem 4.3. Thus 𝑎,𝑏 ∈ 𝜇 𝑡 𝐴 𝑎, 𝑏 ⊆ 𝜇 𝑡 .
µ( ( x * z ) , q ) > max{ µ (x * ( y * z) , q), µ(y,q)} 𝜇 𝑡 = 𝑎,𝑏 ∈ 𝜇 𝑡 𝐴 𝑎, 𝑏 .
1
Taking t1 = { µ((x * z ) , q) + max {µ (x * ( y * z)
2
, q), µ (y,q)}} and µ((x * z ) , q) > t1 > max{µ (x * ( y 5.Anti Q- Fuzzy Sub algebra of KU algebra
* z) , q), µ(y,q)} . Definition 5.1 [11]
It follows that ( x * ( y * z)),y ∈ µt1 and x * z ∉ µt1. A nonempty subset S of a KU-algebra X is
This is a contradiction. said to be KU-subalgebra of X , if x , y S implies
Hence µ((x * z), q) ≤ max{µ (x * ( y * z) , q), µ(y,q)} x * y S.
Therefore µ is an anti Q-fuzzy KU-ideal of X.
Definition 5.2
Definition 4.2 [13] A Q-fuzzy set in a KU-algebra X is called
Let X be an KU- algebra and a,b ∈ X.We can a Q-fuzzy Subalgebra of X if (x * y , q ) ≥ min{(x
define a set A(a,b) by A(a,b) ={ x ∈ X / a* ( b* x) = 0 }. , q), ( y , q)},for all x,y X and q ∈ Q.
It is easy to see that 0,a,b ∈ A(a,b) for all a,b ∈ X.
Definition 5.3 [15]
Theorem 4.3 A Q-fuzzy set in a KU-algebra X is called
Let µ be a Q-fuzzy set in KU-algebra X.Then
an Anti Q-fuzzy sub algebra of X if (x * y , q) ≤
µ is an anti Q- fuzzy KU- ideal of X iff µ satisfies the
max{(x , q), (y , q)}, for all x,y X and q ∈ Q.
following condition.
Example 5.1
∀ 𝑎, 𝑏 ∈ 𝑋 , ∀ 𝑡 ∈ 0,1
From example 3.1 , Let t0, t1, t2 ∈ [ 0, 1] be
𝑎, 𝑏 ∈ 𝜇 𝑡 A a, b ⊆ 𝜇 𝑡
such that t0 < t1 < t2 .Define a Q-fuzzy set µ : X x Q
Proof:
Assume that µ is an anti Q-fuzzy KU- ideal [0, 1] by µ(0,q) = t0, µ(1,q) = t1 = µ(2,q),
of X. µ(3,q) = t2 = µ(4,q) ,By routine calculations µ is an
Let a,b ∈ µt. Then µ (a,q ) ≤ t and µ (b,q ) ≤ t. anti Q-fuzzy subalgebra of X where q∈Q.
Let x ∈ A (a,b). Then a * ( b * x ) = 0. Theorem 5.1
Now, If is an anti Q-fuzzy sub algebra of a KU-
µ (x,q) = µ (0 * x , q) algebra X, then (0,q) ≤ (z,q) , for any z X and q
≤ max { µ (0 * (b * x) , q) , µ ( b,q )} ∈ Q.
= max { µ ((b * x) , q) , µ ( b,q )} Proof
≤ max { max {µ (b * (a * x) , q) , µ (a,q)} , µ(b,q) } Since z * z = 0 for any x X and q ∈ Q, then
= max { max {µ (a * (b * x) , q) , µ (a,q)} , µ(b,q) } ( 0, q ) = ( z * z, q )
= max { max {µ (0,q) , µ (a,q)} , µ(b,q) } ≤ max { ( z, q ) , ( z, q )} = (z,q)
= max { µ (a,q)} , µ(b,q) }≤ max { t , t} = t ( 0, q ) ≤ ( z, q ).
1288 | P a g e
4. P.M.Sitharselvam , T.Priya, T.Ramachandran/ International Journal of Engineering Research
and Applications (IJERA) ISSN: 2248-9622 www.ijera.com
Vol. 2, Issue4, July-August 2012, pp.1286-1289
Theorem 5.2 Conclusion
A Q-fuzzy set µ of a KU– algebra X is an In this article we have discussed anti Q-fuzzy
anti Q-fuzzy subalgebra if and only if for every KU- ideal of KU-algebras and its lower level cuts in
t [0,1] , µt is either empty or a KU-sub algebra of X. detail.These concepts can further be generalized.
Proof: REFERENCES :
[1] Biswas R. , Fuzzy subgroups and Anti Fuzzy
Assume that µ is an anti Q-fuzzy sub algebra of X and subgroups , Fuzzy sets and systems , 35
µt ≠ . Then for any x,y µt and q ∈ Q, we have (1990),121-124.
µ(x * y,q) ≤ max { (x,q), (y,q)} ≤ t. [2] Dudek W.A,On proper BCC-algebras,Bull.
Therefore x*y µt.Henceµt is a KU-sub algebra of X. Inst.Math.Academia Sinica,20(1992),137-
Now Assume that µt is a KU-sub algebra of X. 150.
Then for any x,y µt implies x* y µt . [3] Dudek W.A and Y.B.Jun, Fuzzification of
Take t = max { (x,q), (y,q)}. ideals in BCC- algebras , Glasnik
Therefore µ(x * y,q) ≤ t = max { (x,q), (y,q)}. Matematicki,36 ,(2001) , 127-138.
Hence µ is an Anti Q-fuzzy sub algebra of X. [4] Hu Q.P. and X.Li,On BCH-algebras,
Mathematics Seminar notes11(1983) , 313-
Theorem 5.3 320.
Any KU-sub algebra of a KU– algebra X can [5] Hu Q.P. and X.Li , On Proper BCH-algebras,
be realized as a level KU-sub algebra of some Anti Mathe Japonica 30(1985),659 – 661.
Q-fuzzy sub algebra of X. [6] Iseki K. and S.Tanaka , An introduction to
the theory of BCK – algebras , Math
Proof: Japonica 23 (1978), 1- 20 .
[7] Iseki K., On BCI-algebras , Math.Seminar
Let A be a KU-sub algebra of a given KU – Notes 8 (1980), 125-130.
algebra X and let µ be a Q-fuzzy set in X defined by [8] Muthuraj R.,Sitharselvam P.M,Muthuraman
𝑡 , 𝑖𝑓 𝑥 ∈ 𝐴 M.S., Anti Q – fuzzy group and its lower
µ (x,q) = level subgroups, IJCA,3 (2010) , 16-20.
0 , 𝑖𝑓 𝑥 ∉ 𝐴
where t[0,1] is fixed and q ∈ Q. It is clear that µt =A. [9] Muthuraj .R ,Sridharan M.,Sitharselvam
Now we prove such defined µ is an anti Q- P.M., M.S.Muthuraman,Anti Q- Fuzzy BG-
fuzzy sub algebra of X. ideals in BG-algebra, IJCA, 4 (2010), 27-31.
Let x,y X. If x,y A, then x * y A. [10] Neggers. J , S.S.Ahn and H.S.Kim , On Q-
algebras, IJMMS 27(2001) , 749-757.
Hence ( x, q ) = ( y, q ) = ( x * y , q ) = t and
[11] Prabpayak C. and Leerawat U.,On Ideals and
µ( x * y , q ) ≤ max{ (x,q), (y,q)}= t
congurences in KU-algebras,Scientia Magna
If x,y A, then (x,q) = (y,q) = 0 and J.,5(1)(2009),54-57.
µ(x * y) ≤ max { (x),(y)} = 0. [12] Ramachandran T.,Priya T.,Parimala M., Anti
If at most one of x,y A, then at least one of fuzzy T ideals of TM-algebras and its lower
(x,q) and (y,q) is equal to 0. level cuts, International Journal of Computer
Therefore, max{ (x,q), (y,q)}= 0 so that Applications , volume 43(22),(2012),17-22.
µ(x * y) ≤ 0, which completes the proof. [13] Priya T.,Sithar Selvam P.M., Ramachandran
T, Anti Fuzzy Ideals of CI- algebras and its
Theorem 5.4 lower level cuts, International journal of
Two lower level KU-sub algebras µs, µt (s< t) Mathematical archive, Communicated.
of an anti Q-fuzzy sub algebra are equal iff there is no [14] Samy M.Mostafa ,Mokthar A.Abd-Elnaby
x X and q ∈ Q such that s ≤ µ(x,q) < t. and Moustafa M.M.Yousef, Fuzzy ideals of
KU- algebras, Int. Math. Forum, 6(63) ,
Proof (2011), 3139-3149.
Let µs = µt for some s < t.If there exists xX [15] Sithar Selvam P.M.,Priya T.and
and q ∈ Q such that s ≤ µ(x) < t, then µt is a proper Ramachandran T, Anti Fuzzy Subalgebras
subset of µs, which is a contradiction. and Homomorphism of CI- algebras,
Conversely, assume that there is no xX and q∈Q International journal of Engineering
such that s ≤ µ(x,q) < t. Research & Technology, Accepted for
If x µs, then µ(x,q) ≤ s and µ(x,q) ≤ t. publication.(july 2012,1(5))
Since µ(x,q) does not lie between s and t. [16] Zadeh.L.A. , Fuzzy sets , Inform.control,8
Thus x µt , which gives µs µt , Also µt µs. (1965) , 338 -353.
Therefore µs = µt.
1289 | P a g e