Sie sind auf Seite 1von 6

JOURNAL OF COMPUTING, VOLUME 3, ISSUE 1, JANUARY 2011, ISSN 2151-9617

HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 21

Neutrosophic Logic – Generalization of Vague Relation


a b c
Ashit Kumar Dutta *, Ranjit Biswas , Nasser Saad AL-Arifi
a
Department of Computer Science and Information System
Faculty of Information Technology
SHAQRA University ,Shaqra , Saudi Arabia

b
Department of Computer Science and Information Technology
ITM University, Gurgaon ,Haryana ,India

c
Department of Geophysics, College of Science
King Saud University, Riyadh, Saudi Arabia

ABSTRACT
In this paper, we present a generalization of the Neutrosophic logic [2] over Vague Relation[1]. Our objective is
capable of manipulating incomplete as well as inconsistent information. Fuzzy relation or vague relation can only
handle incomplete information. Each relation is Associated with two membership functions one is called truth-
membership function T which keeps track of the “evidence for you” to which we believe the tuple is in the relation,
another is called falsity-membership function F which keeps track of the “evidence against you” to which we believe
the tuple is not in the relation. A neutrosophic logic is inconsistent if there exists one tuple _ such that T(α) + F(α) > 1 .
In order to handle inconsistent situation, we define the neutrosophic relational model using algebraic operators that are
generalizations of the usual operators such as intersection, union, differentiation, complement on vague relations. Our
data model can underlie any database and knowledge-base management system that deals with incomplete and
inconsistent information.

KEYWORDS:
Vague Relation, Neutrosophic Relation , Vague Set , Neutrosophic Set, Neutrosophic Logic

1. INTRODUCTION neutrosophic set theory which is an extension


of Vague set theory and is capable of
Relational data model was proposed by Ted manipulating incomplete as well as
Codd’s pioneering paper [2]. Since then, inconsistent
relational database systems have been information. We use both truth-membership
extensively studied and a lot of commercial function grade α and falsity-membership
relational database systems are currently
available [3, 4]. This data model usually takes function grade β to denote the status of a tuple
care of only well-defined and unambiguous of a certain relation with ,[0,1]
data. However, when we talk about the
imprecise data or imperfect information ,It will ,2. NRDM is the generalization of
fail to answer. But our Lay users may or may Vague relational data model (VRDM). That is ,
not be aware of imprecision . In order to when α + β = 1, neutroshophic relation is the
represent and manipulate various forms of ordinary Vague relation. We define algebraic
incomplete information in relational databases, operators over i-v neutrosophic logic that
several extensions of the classical relational extend the standard operators such as selection,
model have been proposed [6, 7, 8, 9, 10, 11]. join, union over fuzzy relations.
The vague set and vague logic proposed by
Gau and Behurer provide a requisite 2. VAGUE RELATION AND THEIR
mathematical framework for dealing with PROPERTIES
incomplete and imprecise information. The Fuzzy relations [16] have a wide range of
Vague sets consider both truthmembership T applications in different areas in Computer
and falsity-membership F with and Science, in Management Science, in Banking
T(a),F(a)[0,1] and T(a)+F(a)1.Because of and Finance, in Social Sciences etc.
the restriction it cannot handle inconsistent Our work in this paper is based on the theory
data .In this paper, we present a new relational of relations, specially vague relations. In this
data model neutrosophic relational data model section we recollect the recent literature
(NRDM). Our model is based on the [14,15] on the notion of vague relations and
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 1, JANUARY 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 22

their properties. First of all we mention the R-relation with the element p of Y with the
following notations on interval arithmetic following estimation :-
which will be used in our work here strength of existence of the relation = .6
subsequently. strength of non-existence of the relation = .3
Notations A relation E(X→Y) is called a Complete
Let I[0,1] denotes the family of all closed
Relation from the universe X to the universe Y
subintervals of [0,1]. If I1 = [a1,b1] and I2 =
[a2,b2] be two elements of I[0,1], we call I1 ≥ I2 if VE(x,y) = [1,1] (x,y) X×Y. A relation
if a1 ≥ a 2 and b1≥ b 2 . Similarly we understand (X→Y) is called a Null Relation from the
the relations I1 ≤ I2 and I1 = I2 . Clearly the
universe X to the universe Y if V(x,y) =
relation I1≥I2 does not necessarily imply that I1
[0,0] (x,y) X×Y.
I2 and conversely. Also for any two unequal
intervals I1 and I2, there is no necessity that
2.2 Various Operations on VRs
either I1≥ I2 or I1 ≤ I2 will be true. The term
‘imax’ means the maximum of two intervals as
For suitable applications of vague relations, we
imax(I1,I2) = [max(a1,a2), max(b1,b2)].
must be aware of the different operations on
Similarly defined is ‘imin’. The concept of
them. In this section we define some
‘imax’ and ‘imin’ could be extended to define
operations on VRs.
‘isup’ and ‘iinf’ of infinite number of elements
of I[0,1].
Definition 2.2.1 Complement of a VR
It is obvious that L = { I[0,1], isup, iinf,≤ } is a
lattice with universal bounds [0,0] and [1,1]. Let R(X→Y) be a VR describing some relation
R. Its complement denoted by Rc(X→Y) is the
VR given by VRc(x,y) = [ fR(x,y), 1–tR(x,y) ]
2.1 Vague Relation (VR)
Definition 2.2.2 Union of two VRs
Let X and Y be two universes. A vague Let R(X→Y) and S(X→Y) be two VRs each
from the universe X to the universe Y. The
relation (VR) denoted by R(X→Y) of the
union of R and S is denoted by RS which is
universe X with the universe Y is a VS of the
Cartesian product X×Y. also a VR from X to Y, and is given by VRS
(x,y) = imax { VR(x,y), VS(x,y) }.
The true membership value tR(x,y) estimates
the strength of the existence of the relation of Definition 2.2.3 Intersection of two VRs
R-type of the object x with the object y, Let R(X→Y) and S(X→Y) be two VRs each
whereas the false membership value fR(x,y) from the universe X to the universe Y. The
estimates the strength of the non-existence of
intersection of R and S is denoted by RS
the relation of R-type of the object x with the
which is also a VR from X to Y, and is given
object y. The relation R(X→Y) could be in by
short denoted by the notation R, if there is no VRS (x,y) = imin { VR(x,y), VS(x,y) }.
confusion. The following proposition is straightforward.
Example :
Consider two universes X = {a,b} and Y = Proposition 2.2.1
{p,q,r}. Let R be a VR of the universe X with
the universe Y proposed by an intelligent agent
Let R(X→Y), S(X→Y) and T(X→Y) be three
as shown by the following table :-
VRs each relating the universe X with the
VR R(X→Y) universe Y. Then
R(X→Y) P q r (i) (Rc)c = R
(ii) RS = SR
(iii) RS = SR
X (. 6, . 3) (. 3, . 5) ( . 8,. 2)
(iv) R(S T) = (RS) T
Y (. 2 , . 4) (. 7, . 3) ( . 4 , . 4) (v) R(S T ) = (RS) T
(vi) R(S T) = (RS) (R T)
(vii) R(S T ) = (RS) (R T )
The proposed VR reveals the strength of vague (viii) RRc ≠ E and RRc≠ 
relation of every pair of X×Y; For example, it 
reveals that the object y of the universe X has
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 1, JANUARY 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 23

De Morgans Laws hold good in the operations 4. Neutrosophic Logic Connectors


with VRs. One uses the definitions of neutrosophic
probability and neutrosophic set operations.
Proposition 2.2.2 Similarly, there are many ways to construct
Let R(X→Y) and S(X→Y) be two VRs each such connectives according to each particular
problem to Solve; here we present the easiest
relating the universe X with the universe Y.
ones:
Then the following results are true:-
One notes the neutrosophic logic values of the
(i) (RS)c = RcSc propositions A1 and A2 by NL(A1) = ( T1, I1,
(ii) (RS)c = RcSc F1 ) and NL(A2) = ( T2, I2, F2 ) respectively.
For all neutrosophic logic values below: if,
3. THEORY OF NEUTROSOPHIC SETS after
calculations, one obtains numbers < 0 or 1>,
In this section, we present the basic operations one
of replaces them
Neutrosophic sets which include complement, -
0 or 1+ respectively.
containment, equal, union, intersection, and so
on. 4.1. Negation:
Let S1 and S2 be two (unidimensional) real
NL(A1) = ( {1+}ӨT1, {1+}ӨI1, {1+}ӨF1 )
standard or non-standard subsets, then one
4.2. Conjunction:
defines:
NL(A1 A2) = ( T1T2, I1I2, F1F2 ).
3.1 Addition of Sets: (And, in a similar way, generalized for n
propositions.)
S1S2 = {x|x=s1+s2, where s1S1 and
s2S2} with inf S1S2 = inf S1 + inf S2, sup 4.3 Implication:
S1S2 = sup S1 + sup S2; NL(A1 A2) = ( {1+}ӨT1T1T2,
and, as some particular cases, we have {a}S2 {1+}ӨI1I1I2, {1+}ӨF1F1F2 )
= {x|x=a+s2, where s2S2} with inf {a}S2
= a + inf S2, sup {a}S2 = a + sup S2. 5. NEUTROSOPHIC LOGIC
In this section, we generalize vague relations
3.2 Subtraction of Sets: in such a manner that we are now able to
assign a measure of truth ness and a measure
S1ӨS2 = {x|x=s1-s2, where s1S1 and
of falseness to each tuple. We shall refer to
s2S2}. these generalized vague relations as
For real positive subsets (most of the cases will neutrosophic logic. So, a tuple in a
fall in this range) one gets inf S1ӨS2 = inf S1 - neutrosophic logic is assigned a measure. will
sup S2, sup S1ӨS2 = sup S1 – inf S2; be referred to as the truth factor and will be
and, as some particular cases, we have referred to as the false factor.
{a}ӨS2 = {x|x=a-s2, where s2S2}, with inf The interpretation of this measure is that we
{a}ӨS2 = a - sup S2, sup {a}ӨS2 = a - inf believe with confidence and doubt with
S2; confidence that the tuple is in the relation. The
truth and false confidence factors for a tuple
3.3 Multiplication of Sets: need not add to exactly 1. This allows for
S1S2 = {x|x=s1.s2, where s1S1 and incompleteness and inconsistency to be
represented. If the truth and false factors add
s2S2}. up to less than 1, we have incomplete
For real positive subsets (most of the cases will information regarding the tuple’s status in the
fall in this range) one gets inf S1S2 = inf S1 relation and if the truth and false factors add up
. inf S2, sup S1S2 = sup S1sup S2; to more than 1, we have inconsistent
and, as some particular cases, we have information regarding
{a}S2 = {x|x=as2, where s2S2}, with inf the tuple’s status in the relation.
In contrast to vague relations where the grade
{a}S2 = a * inf S2, sup {a}S2 = asup of membership of a tuple is fixed, neutrosophic
S2; logic bound the grade of membership of a
3.4 Division of a Set by a Number: tuple to a subinterval [,1]for the
Let k R* then S1/k = {x|x=s1/k, where case,1.. The operators on fuzzy
relations can also be generalized for
s1S1}.
neutrosophic logic. However, any such
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 1, JANUARY 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 24

generalization of operators should maintain the R1,….Rn on schemes ∑ 1,…., ∑ n+1, respectively.
belief system intuition behind neutrosophic s. (R1,….Rn) is also total.
Definition 5.1 A neutrosophic relation on Definition 5.4
scheme R on ∑ is any subset of (∑)×[0,1] A totality preserving operator  on
×[0,1], Where (∑) denotes the set neutrosophic logic with signature〈∑ 1,………,
of all tuples on any scheme ∑ . ∑ n+1 〉 is a weak generalization of an
For any t (∑) ,we shall denote an element operator on fuzzy relations with the same
of R as 〈 t,R(t)+,R(t)-〉 ,where , R(t)+ is the signature, if for any total neutrosophic logic
truth factor assigned to t by R and R(t)- is the R1,….Rn on schemes ∑ 1,….,∑ n, respectively
,we have
false factor assigned to t by R.Let V(∑ ) be the
set of all neutrosophic relation on ∑ . ∑ n+1(R1,….Rn))= ∑
( 1(R1),…..,

Definition 5.2 A neutrosophic relation on ∑ n(Rn)).

scheme R on ∑ is consistent if R(t)++R(t)- 1, The above definition essentially requires  to


coincide with on total neutrosophic realtions
for all t (∑) .Let C(∑ )be the set of all (which are in One-one correspondence with
consistent neutrosophic logic on∑ .R is said to the vague relations). In general, there may be
be complete if R(t)++R(t)- 1, for all t (∑) many operators on neutrosophic logic that are
. If R is both consistent and complete, i.e. weak generalizations of a given operator on
fuzzy relations. The behavior of the weak
R(t)++R(t)- = 1, for all t (∑) . then it is a
total generalizations of on even just the
neutrosophic relation, and let T(∑ ) be the set consistent neutrosophic logic may in general
of total neutrosophic relation on ∑ . vary. We require a stronger notion of operator
generalization under which, at least when
5.1 Operator Generalizations restricted to consistent neutrosophic logic, the
It is easily seen that neutrosophic logic are a behavior of all the generalized operators is the
generalization of vague relations, in that for same. Before we can develop such a notion,
each we need that of ‘representation’ of a
vague relation there is a neutrosophic relation neutrosophic relation.
with the same information content, but not vice We associate with a consistent neutrosophic
versa. It is thus natural to think of generalizing relation R the set of all (vague relations
the operations on vague relations such as corresponding to) total neutrosophic logic
union, join, and projection etc. to neutrosophic obtainable from R by filling the gaps between
logic. the truth and false factors for each tuple.
However, any such generalization should be Let the map be reps ∑ :C(∑ )→2F( ∑ ).is given
intuitive with respect to the belief system by , reps ∑ (R)={QF(∑ )| (R(ti)+ Q(ti)
model of neutrosophic logic. We now 1-R(ti)- )}.
construct a
ti(∑ )
The set reps ∑ (R) contains all fuzzy relations
framework for operators on both kinds of that are ‘completions’ of the consistent
relations and introduce two different notions of neutrosophic relation R .Observe that reps∑ is
the generalization relationship among their defined only for consistent neutrosophic logic
operators. An n-ary operator on fuzzy relations and produces sets of fuzzy relations. Then we
with signature have following observation.
〈∑ 1,………, ∑ n+1〉 is a function:F(∑ 1) Proposition 5.1 For any consistent
×…×F(∑ n)→∑ n+1,where ∑ 1,…., ∑ n+1 are any neutrosophic
schemes. Similarly An n-ary operator on relation R on scheme ∑ , reps ∑ (R) is the
neutrosophic logic with signature〈∑ 1,………, singleton
∑ n+1〉 is a function :V(∑ 1) 
{ ∑ ( R)}, iff R is total.
×…×V(∑ n)→V(∑ n+1). Proposition 5.2 If is a strong generalization
Definition 5.3 of , then  is also a weak generalization of 
An operator on neutrosophic logic with
signature 〈∑ 1,………, ∑ n+1 〉 is totality
preserving if for any total neutrosophic logic
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 1, JANUARY 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 25

relations, such as union, intersection,


6. APPLICATION complement, differentiation, we introduced
In the real world there are vaguely specified generalized operators on neutrosophic logic.
data values in many applications, such as These generalized operators maintain the belief
sensor system intuition behind neutrosophic logic,
information,Robotics etc. A Vague Set (VS), and are shown to be “well-behaved” in the
as well as an Intuitionistic Fuzzy Set (IFS), is a sense
further generalization of an FS. Now take an mentioned above. Our data model can be used
example, when we ask the opinion of an expert to
about certain statement, he or she may say that represent relational information that may be
the possibility that the statement is true is incomplete and inconsistent. As usual, the
between 0.5 and 0.7, and the statement is false algebraic operators can be used to construct
is between 0.2 and 0.4, and the degree that he queries to any database systems for retrieving
or she is not sure is between 0.1 and 0.3. Here vague information.
is another example, suppose there are 10 voters REFERENCES
during a voting process. In time t1, three vote [1] Haibin Wang, Praveen Madiraju, Yang-
“yes”, two vote “no” and five are undecided, Qing Zhang, and Rajshekhar Sunderraman,
using neutrosophic notation, it can be Interval Neutrosophic Sets, International
expressed as x(0.3,0.5,0.2); in time t2, three Journal of Applied Mathematics & Statistics,
vote “yes”, two vote “no”, two give up and vol. 3, no. M05, pp. 1-18, March 2005.
three are undecided, it then can be expressed as [2] F. Smarandache (2002a), A Unifying Field
x(0.3,0.3,0.2). That is beyond the scope of the in Logics: Neutrosophic Logic, in Multiple-
intuitionistic fuzzy set. Valued Logic / An International Journal, Vol.
So, the notion of neutrosophic set is more 8, No.3,385-438,2002,
general and overcomes the aforementioned [3] E.F. Codd, A Relational Model for Large
issues. In neutrosophic set, indeterminacy is Shared Data Banks, Communications of the
quantified explicitly and truth membership, ACM, 13(6):377- 387, June 1970.
indeterminacy-membership and falsity [4] Elmasri and Navathe, Fundamentals of
membership are independent. This assumption DatabaseSystems, Addison-Wesley,New York,
is very third edition, 2000.
important in many applications such as [5] A. Silberschatz, H. F. Korth, and S.
information fusion in which we try to combine Sudarshan, Database System Concepts,
the data from different sensors. Neutrosophy MCGraw-Hill, Boston, third edition, 1996.
was introduced by Smarandache[28] . [6] S. Parsons, Current Approaches to Handing
Neutrosophic set is a powerful general formal Imperfect Information in Data and Knowledge
framework which generalizes the concept of Bases, IEEE Trans, Knowledge and Data
the classic set, fuzzy set , Vague set etc.A Engineering, 3:353- 372, 1996.
neutrosophic set A defined on universe U. x = [7] J. Biskup, A Foundation of Codd’s
x(T,I,F) ε A with T,I and F being the real Relational
standard or non-standard Maybe-operations, ACM Trans. Database
subsets of ]0-,1+[, T is the degree of truth- Systems, 8, 4:608-636, Dec. 1983.
membership of A, I is the degree of [8] M. L. Brodie, J. Mylopoulous, and J. W.
indeterminacy membership of A and F is the Schmidt, On the Development of Data Models,
degree of falsity-membership of A. On Conceptual Modeling, 19-47, 1984.
[9] E. F. Codd, Extending the Database
Relational Model to Capture More Meaning,
7. CONCLUSION ACM Trans. Database Systems, 4(4):397-434,
We have presented a generalization of Vague Dec. 1979.
relations, called neutrosophic logic, in which [10] W. Lipski, On Semantic Issues Connected
we allow the representation of three factors with Incomplete Information Databases, ACM
Truth, False, Indeterminacy with each tuple. Trans. Database Systems, 4, 3:262-296, Sept.
The algebra on vague relations is appropriately 1979.
generalized to manipulate [11] W. Lipski, On Databases with Incomplete
neutrosophic logic. We developed two notions Information, Journal of the Association for
of generalizing operators on Vague relations Computing Machinery, 28:41-70, 1981.
for neutrosophic logic. Of these, the stronger [12] D. Maier, The Theory of Relational
notion guarantees that any generalized operator Databases, Computer Science Press,
is “wellbehaved” for neutrosophic relation Rockville, Maryland, 1983.
operands that contain consistent information. [13] K. C. Liu and R. Sunderraman, Indefinite
For some well-known operators on vague and Maybe Information in Relational
JOURNAL OF COMPUTING, VOLUME 3, ISSUE 1, JANUARY 2011, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 26

Databases, ACM Trans. Database Systems, databases”, Fuzzy Sets and Systems 92, pp.1-
15(1):1-39, 1990. 10.
[14] K. C. Liu and R. Sunderraman, A
Generalized Relational Model for Indefinite
and Maybe Information, IEEE Transaction on
Knowledge and Data Engineering, 3(1):65-77,
1991.
[15] L. A. Zadeh, Fuzzy Sets, Inf. Control,
8:338-353, 1965.
[16] I. Turksen, Interval Valued Fuzzy Sets
Based on Normal Forms, Fuzzy Sets and
Systems, 20:191-210, 1986. 24
[17] K. Atanassov, Intuitionistic Fuzzy Sets,
Fuzzy Sets and Systems, 20:87-96, 1986.
[18] M. Anvari and G. F. Rose, Fuzzy
Relational
Databases, In Proceedings of the 1st
International Conference on Fuzzy
Information Processing, Kuaui, Hawaii, CRC
Press, 1984.
[19] J. F. Baldwin, A Fuzzy Relational
Inference
Language for Expert Systems, In Proceedings
of the 13th IEEE International Symposium on
Multivalued Logic, Kyoto, Japan, 416-423,
1983.
[20] J. Kacprzyk and A. Ziolkowski, Database
Queries with Fuzzy Linguistic Quantifier,
IEEE Trans. Syst. Man Cyber. 16, 3:474-479,
May/June, 1986.
[21] L. A. Zadeh, Fuzzy Sets as the Basis for a
Theory of Possibility, Fuzzy Sets and Systems,
1:1-27, 1978.
[22] H. Prade and C. Testemale, Generalizing
Database Relational Algebra for the Treatment
of Incomplete or Uncertain Information and
Vague Queries, Information Sciences, 34:115-
143, 1984.
[23] H. Prade and C. Testemale,
Representation of Soft Constraints and Fuzzy
Attribute Values by Means of Possibility
Distributions in Databases, Analysis of Fuzzy
Information, Volume II, Artificial Intelligence
and Decision Systems, 213-229, 1987.
[24] Gau, W.L. and Buehrer, D, J., Vague sets,
IEEE Transactions on Systems, Man and
Cybernetics, Vol.23 (1993) 610-614.
[25] Atanassov,K., Intuitionistic Fuzzy Sets,
Fuzzy Sets and Systems.20(1986):87-96.
[26] Atanassov, K., Intuitionistic Fuzzy Sets :
Theory and Applications, Physica-Verlag
(2000), New- York.
[27] Bustince, H. and Burillo, P., Vague sets
are
intuitionistic fuzzy sets, Fuzzy Sets and
Systems 79 (1996) 403-405.
[28] Chiang D., Chow L. R. and Hsien N,
“Fuzzy information in extended fuzzy
relational

Das könnte Ihnen auch gefallen