You are on page 1of 9

# 80 Neutrosophic Sets and Systems, Vol.

16, 2017

## Rough Neutrosophic Multisets

1
Faculty of Computer and Mathematical Sciences, Universiti Teknologi Mara (UiTM) Kelantan, Campus Machang, Kelantan, 18500, Malaysia. E-
mail: suria588@kelantan.uitm.edu.my
2
Faculty of Computer and Mathematical Sciences, Universiti Teknologi Mara (UiTM) Shah Alam, Shah Alam, 40450, Malaysia. E-mail:
daud@tmsk.uitm.edu.my
3
Faculty of Computer and Mathematical Sciences, Universiti Teknologi Mara (UiTM) Shah Alam, Shah Alam, 40450, Malaysia. E-mail:

Abstract. Many past studies largely described the con- intersection. We also have examined some desired prop-
cept of neutrosophic sets, neutrosophic multisets, rough erties of rough neutrosophic multisets based on these def-
sets, and rough neutrosophic sets in many areas. Howev- initions. We use the hybrid structure of rough set and
er, no paper has discussed about rough neutrosophic mul- neutrosophic multisets since these theories are powerful
tisets. In this paper, we present some definition of rough tool for managing uncertainty, indeterminate, incomplete
neutrosophic multisets such as complement, union and and imprecise information.

Keywords: Neutrosophic set, neutrosophic multiset, rough set, rough neutrosophic set, rough neutrosophic multisets

## 1 Introduction rough set that combines interval- valued neutrosophic sets

and rough sets. It studies roughness in interval- valued
In our real-life problems, there are situations with un-
neutrosophic sets and some of its properties. After the in-
certain data that may be not be successfully modelled by
troduction of rough neutrosophic set theory, many interest-
the classical mathematics. For example, the opinion about
ing application have been studied such as in medical or-
beauty, which is can be describe by more beauty, beauty,
ganisation , , .
beauty than, or less beauty. Therefore, there are some
But until now, there have been no study on rough neu-
mathematical tools for dealing with uncertainties such as
trosophic multisets (RNM). Therefore, the objective of this
fuzzy set theory introduced by Zadeh , intuitionistic
paper is to study the concept of RNM which is combina-
fuzzy set theory introduced by Attanasov , rough set
tion of rough set  and neutrosophic multisets  as a
theory introduced by Pawlak , and soft set theory initi-
generalization of rough neutrosophic sets .
ated by Molodtsov . Rough set theory introduced by
This paper is arranged in following manner. In section
Pawlak in 1981/1982, deals with the approximation of sets
2, some mathematical preliminary concepts were recall for
that are difficult to describe with the available information.
more understanding about RNM. In section 3, the concepts
It is expressed by a boundary region of set and also ap-
of RNM and some of their properties are presented with
proach to vagueness. After Pawlaks work several re-
examples. Finally, we conclude the paper.
searcher were studied on rough set theory with applications
, .
2 Mathematical Preliminaries
However, these concepts cannot deal with indetermi-
nacy and inconsistent information. In 1995, Smarandache In this section, we mainly recall some notions related
 developed a new concept called neutrosophic set (NS) to neutrosophic sets , , , neutrosophic multisets
which generalizes probability set, fuzzy set and intuition- , , , , rough set , and rough neutrosophic
istic fuzzy set. There are three degrees of membership de- set , , that relevant to the present work and for fur-
scribed by NS which is membership degree, indeterminacy ther details and background.
degree and non-membership degree. This theory and their
hybrid structures has proven useful in many different field
, , , ,, ,.
Broumi et al.  proposed a hybrid structure called
neutrosophic rough set which is combination of neutro-
sophic set  and rough set  and studied their properties.
Later, Broumi et al.  introduced interval neutrosophic

Neutrosophic Sets and Systems, Vol. 16, 2017 81

Definition 2.1 (Neutrosophic Set)  Let X be an uni- to explain the characteristics of the objects. Suppose A is a
verse of discourse, with a generic element in X denoted by neutrosophic set (NS) of U, such that,
x, the neutrosophic (NS) set is an object having the form
A = {x1, (0.3, 0.4, 0.5) , x2, (0.5, 0.1, 0.4) ,
A = {x, (TA(x), IA(x), FA(x)) | x X} x3, (0.4, 0.3, 0.5)},
where the degree of goodness of prices is 0.4, degree of in-
where the functions T, I, F : X ]0, 1+[ define respective-
determinacy of prices is 0.3 and degree of poorness of
ly the degree of membership (or Truth), the degree of inde-
prices is 0.5 etc.
terminacy, and the degree of non-membership (or False-
hood) of the element x X to the set A with the condition The following definitions are refer to .
Definition 2.3 (Neutrosophic Multisets) Let E be a

0 TA(x) + IA(x) + FA(x) 3+
universe. A neutrosophic multiset (NMS) A on E can be
defined as follows:
From a philosophical point of view, the neutrosophic set
takes the value from real standard or non-standard subsets A x, (TA1 ( x),TA2 ( x),...,TAp ( x)),( I 1A ( x), I A2 ( x),...,I Ap ( x)),
of ]0, 1+[. So, instead of ]0, 1+[ we need to take the inter-
val [0, 1] for technical applications, because ]0, 1+[ will be ( FA1 ( x), FA2 ( x),...,FAp ( x)) : x E
difficult to apply in the real applications such as in scien-
tific and engineering problems. Therefore, we have
where, the truth membership sequence
A = {x, (TA(x), IA(x), FA(x)) | x X, TA(x), IA(x), (TA1 ( x),TA2 ( x),...,TAp ( x)) , the indeterminacy membership
FA(x) [0, 1]}. sequence ( I 1A ( x), I A2 ( x),...,I Ap ( x)) and the falsity membership
sequence ( FA1 ( x), FA2 ( x),...,FAp ( x)) may be in decreasing or
There is no restriction on the sum of TA(x); IA(x) and FA(x),
so increasing order, and the sum of

## 0 TA(x) + IA(x) + FA(x) 3

TAi ( x), I Ai ( x), FAi ( x) 0,1 satisfies the condition
0 TAi ( x) I Ai ( x) FAi ( x) 3
for any x E and i =1, 2,, p.
Also, p is called the dimension (cardinality) of NMS A.
For two NS,
For convenience, a NMS A can be denoted by the sim-
A = {x, (TA(x), IA(x), FA(x)) | x X} and plified form:

## B = {x, (TB(x), IB(x), FB(x)) | x X} A x, (TAi ( x), I Ai ( x), FAi ( x) x E , i 1,2,..., p}

the relations are defined as follows: Definition 2.4 Let A,B NMS(E). Then,
(i) A B if and only if TA(x) TB(x), IA(x) IB(x),
FA(x) FB(x), ~B
(i) A is said to be NM subset of B is denoted by A
(ii) A = B if and only if TA(x) = TB(x), IA(x) = IB(x),
if TAi ( x) TBi ( x) , I Ai ( x) I Bi ( x) ,
FA(x) = FB(x),
(iii) A B = {x, min(TA(x),TB(x)), max(IA(x),IB(x)), FAi ( x) FBi ( x) , x E .
max(FA(x),FB(x)) | x X},
(iv) A B = {x, max(TA(x),TB(x)), min(IA(x),IB(x)), (ii) A is said to be neutrosophic equal of B is denoted
min(FA(x),FB(x)) | x X}, by A = B if
(v) Ac = {x, FA(x), 1 IA(x), TA(x) | x X}
(vi) 0n = (0, 1, 1) and 1n = (1, 0, 0). TAi ( x) TBi ( x) , I Ai ( x) I Bi ( x) ,

## As an illustration, let us consider the following example.

FAi ( x) FBi ( x) , x E.
~

Example 2.2. Assume that the universe of discourse U = (iii) The complement of A denoted by AC is defined by
{x1, x2, x3}, where x1 characterizes the capability, x2 char- ~
AC x, ( FA1 ( x), FA2 ( x),...,FAp ( x)),
acterizes the trustworthiness and x3 indicates the prices of
the objects. It may be further assumed that the values of x1, (1 I 1A ( x),1 I A2 ( x),...,1 I Ap ( x)),
x2, and x3 are in [0, 1] and they are obtained from some (TA1 ( x), TA2 ( x),...,TAp ( x)) : x E
questionnaires of some experts. The experts may impose
their opinion in three components which is the degree of
goodness, the degree of indeterminacy and that of poorness

82 Neutrosophic Sets and Systems, Vol. 16, 2017

## (iv) If TA ( x) 0 and I Ai ( x) FAi ( x) 1 for all x E

i (iv) The multiplication of A and B is denoted by
A~ B H and is defined by
and i 1, 2, ..., p, then A is called null ns-set and
~ H x, (TH1 ( x),TH2 ( x),...,THp ( x)),
denoted by .
( I 1H ( x), I H2 ( x),...,I Hp ( x)),
(iv) If TA ( x) 1 and I Ai ( x) FAi ( x) 0 for all x E
i
( FH1 ( x), FH2 ( x),...,FHp ( x)) | x E
and i 1, 2, ..., p, then A is called universal ns-
~
set and denoted by E . where

Definition 2.5 Let A,B NMS(E). Then, THi ( x) TAi ( x) TBi ( x),
~ I Hi ( x) I Ai ( x) I Bi ( x) I Ai ( x) I Bi ( x),
(i) The union of A and B is denoted by A B C is
defined by FHi ( x) FAi ( x) FBi ( x) FAi ( x) FBi ( x),

## ( I C1 ( x), I C2 ( x),...,I Cp ( x)), Here , , , , denotes minimum, maximum, addition,

multiplication, subtraction of real numbers respectively.
( F ( x), F ( x),...,F ( x)) : x E
1
C C
2
C
p

## Definition 2.6 (Rough Set)  Let R be an equivalence

where relation on the universal set U. Then, the pair (U, R) is
TCi ( x) TAi ( x) TBi ( x), I Ci ( x) I Ai ( x) I Bi ( x) , called a Pawlaks approximation space. An equivalence
class of R containing x will be denotes by [x]R . Now, for X
FCi ( x) FAi ( x) FBi ( x), U, the upper and lower approximation of X with the re-
spect to (U, R) are denoted by, respectively A1(x) and A2(x)
for x E and i 1, 2, ..., p .
and defined by
(ii) The intersection of A and B is denoted by A1(x) = {x : [x]R X} and A2(x) = {x : [x]R X }
~
A B D and is defined by
Now, if A1(x) = A2(x), then X is called definable; otherwise,
D x, (TD1 ( x),TD2 ( x),...,TDp ( x)), the pair A(X) = (A1(x), A2(x)) is called the rough set of X in
( I 1D ( x), I D2 ( x),...,I Dp ( x)),
U.
Example 2.7  Let A =(U, R) be an approximate space
( FD1 ( x), FD2 ( x),...,FDp ( x)) | x E
where U = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10} and the relation R
where on U be definable aRb iff a b (mod 5) for all a, b U.
Let us consider a subset X = {1, 2, 6, 7, 8, 9} of U. Then,
TDi ( x) TAi ( x) TBi ( x), I Di ( x) I Ai ( x) I Bi ( x) ,
the rough set of X is A(x) = ( A (x), A(x)) where A (x) = {1,
FDi ( x) FAi ( x) FBi ( x), 2, 6, 7} and A(x) = {1, 2, 3, 4, 6, 7, 8, 9}. Here, the equiv-
for x E and i 1, 2, ..., p . alence classes are
R = R = R = {0, 5, 10}
(iii) The addition of A and B is denoted by A ~ B G
and is defined by R = R = {1, 6}
G x, (TG1 ( x), TG2 ( x),...,TGp ( x)), R = R = {2, 7}
( I G1 ( x), I G2 ( x),...,I Gp ( x)), R = R = {3, 8}
( FG1 ( x), FG2 ( x),...,FGp ( x)) | xE R = R = {4, 9}
where Thus, A (x) = {x U: [x]R X } ={1, 2, 6, 7} and

TGi ( x) TAi ( x) TBi ( x) TAi ( x) TBi ( x), A(x) = {x : [x]R X }= {1, 2, 3, 4, 6, 7, 8, 9}.

## I Gi ( x) I Ai ( x) I Bi ( x) , The following definitions are refer to .

FGi ( x) FAi ( x) FBi ( x),

## for x E and i 1, 2, ..., p .

Neutrosophic Sets and Systems, Vol. 16, 2017 83

Definition 2.8 Let A = (A1,A2) and B = (B1,B2) be two sophic set (0N) and unit neutrosophic sets (1N) are definable
rough sets in the approximation space S = (U, R). Then, neutrosophic sets. Let consider the example in the follow-
ing.
(i) A B = (A1 B1, A2 B2),
Example 2.10 Let U = { p1, p2, p3, p4, p5, p6, p7, p8 }
(ii) A B = (A1 B1, A2 B2), be the universe of discourse. Let R be an equivalence rela-
(iii) A B if A B = A, tion its partition of U is given by
(iv) A = {U A2, U A1}. U/R = { {p1, p4}, {p2, p3, p6}, { p5}, { p7, p8} }.
Definition 2.9 (Rough Neutrosophic Set) Let U be a
non-null set and R be an equivalence relation on U. Let A Let N(A) = { (p1, (0.3, 0.4, 0.5), (p4, (0.4, 0.6, 0.5)), (p5,
be neutrosophic set in U with the membership function TA, (0.5, 0.7, 0.3)), (p7, (0.2, 0.4, 0.6)) } be a neutrosophic set
indeterminacy function IA and non-membership function of U. By definition 2.6 and 2.9, we obtain:
FA. The lower and the upper approximations of A in the
approximation (U, R) denoted by N (A) and N ( A) are re- N(A) = { (p1, (0.3, 0.6, 0.5)), (p4, (0.3, 0.6, 0.5)), (p5,
spectively defined as follows: (0.5, 0.7, 0.3)) } and

N (A) = {x, (TA(x), IA(x), FA(x)) | y [x]R, x U }, N (A) = { (p1, (0.3, 0.4, 0.5)), (p4, (0.3, 0.4, 0.5)), (p5,
(0.5, 0.7, 0.3)), (p7, (0.2, 0.4, 0.6)), (p8, (0.2, 0.4, 0.6)) }.
N ( A) = {x, (TA(x), IA(x), FA(x)) | y [x]R, x U }

## where For another neutrosophic sets,

TN ( A) ( x) TA ( y ), I N ( A) ( x) I A ( y ), N(B) = { (p1, (0.3, 0.4, 0.5), (p4, (0.3, 0.4, 0.5)), (p5,
yx R yx R (0.5, 0.7, 0.3)) }.
FN ( A) ( x) FA ( y ) The lower approximation and upper approximation of N(B)
yx R
are calculated as
TN ( A) ( x) TA ( y ), I N ( A) ( x) I A ( y ),
yx R yx R N(B) = { (p1, (0.3, 0.4, 0.5)), (p4, (0.3, 0.4, 0.5)), (p5,
(0.5, 0.7, 0.3)) } and
FN ( A) ( x) FA ( y )
yx R
N (B) = { (p1, (0.3, 0.4, 0.5)), (p4, (0.3, 0.4, 0.5)), (p5,
So, (0.5, 0.7, 0.3)) }.
0 TN ( A) ( x) I N ( A) ( x) FN ( A) ( x) 3 , and Obviously, N(B) = N (B) is a definable neutrosophic set in
the approximation space (U, R).
0 TN ( A) ( x) I N ( A) ( x) FN ( A) ( x) 3
Definition 2.11 If N(A) = (N(A), N ( A)) is a rough neutro-
Here and denote min and max operators re- sophic set in (U, R), the rough complement of N(A) is the
spectively. TA(y), IA(y) and FA(y) are the membership, inde- rough neutrosophic set denoted by ~N(A) = (N(A)c,
terminacy and non-membership degrees of y with respect N ( A)c ) where N(A)c, N ( A) c are the complements of neu-
to A. N (A) and N ( A) are two neutrosophic sets in U. trosophic sets N(A) and N ( A) respectively,

## Thus, NS mappings N , N : N U N U are, respec- N ( A) c = { x, FN ( A) ( x),1 I N ( A) ( x),TN ( A) ( x) | x U},

tively, referred to as the upper and lower rough NS approx-
imation operators, and the pair is (N (A), N ( A)) called the N ( A) c = { x, FN ( A) ( x),1 I N ( A) ( x),TN ( A) ( x) | x U}
rough neutrosophic set in (U, R).
Definition 2.12 If N ( F1) and N ( F2 ) are two rough neu-
Based on the above definition, it is observed that N (A) trosophic set of the neutrosophic sets F1 and F2 respective-
and N ( A) have a constant membership on the equivalence ly in U, then we define the following:
classes of U, if N (A) = N ( A) ; i.e.,
(i) N(F1) = N(F2) iff N ( F1 ) N ( F2 ) and
TN ( A) ( x) TN ( A) ( x), I N ( A) ( x) I N ( A) ( x),
N ( F1 ) N ( F2 )
FN ( A) ( x) I N ( A) ( x)
(ii) N(F1) N(F2) iff N ( F1) N ( F2 ) and
For any x U , A is called a definable neutrosophic N ( F1 ) N ( F2 )
set in the approximation (U, R). Obviously, zero neutro-

84 Neutrosophic Sets and Systems, Vol. 16, 2017

## (iii) N ( F1 ) N ( F2 ) N ( F1 ) N ( F2 ), N ( F1 ) N ( F2 ) 3 Rough Neutrosophic Multisets

Based on the equivalence relation on the universe of
(iv) N ( F1 ) N ( F2 ) N ( F1 ) N ( F2 ), N ( F1 ) N ( F2 ) discourse, we introduce the lower and upper approxima-
tions of neutrosophic multisets  in a Pawlaks approx-
imation space  and obtained a new notion called rough
(v) N ( F1) N ( F2 ) N ( F1 ) N ( F2 ), N ( F1 ) N ( F2 ) neutrosophic multisets (RNM). Its basic operations such as
complement, union and intersection also discuss over them
with the examples. Some of it is quoted from ,
(vi) N ( F1) N ( F2 ) N ( F1 ) N ( F2 ), N ( F1 ) N ( F2 )
,, .
If N, M, L are rough neutrosophic set in (U, R), then the re- Definition 3.1 Let U be a non-null set and R be an equiv-
sults in the following proposition are straightforward from alence relation on U. Let A be neutrosophic multisets in U
definitions. with the truth membership sequence TAi , indeterminacy

## Proposition 2.13. membership sequences I Ai and falsity membership se-

quences FAi . The lower and the upper approximations of A
(i) N (~N) = N
in the approximation (U, R) denoted by Nm ( A) and
(ii) N M M N , N M M N
Nm(A) are respectively defined as follows:
(iii) ( N M ) L N (M L), and
Nm ( A) = {x, ( TNm( A) ( x) , I Nm( A) ( x) , FNm( A) ( x) ) |
i i i
( N M ) L N (M L)

i
( A)
( x ) , I Nm
i
( A)
( x ) , FNm
i
( A)
( x) ) |

## De Morgans Laws are satisfied for rough neutrosophic y [x]R, x U},

sets:
where
Proposition 2.14. i =1, 2, .., p,
(i) ( N (F1) N (F2) ) = (N (F1)) (~ N (F2))
( A) ( x ) T Ai ( y ),
i
TNm
yx R
(ii) ( N (F1) N (F2) ) = (N (F1)) (~N (F2))
( A) ( x ) I Ai ( y ),
i
I Nm
yx R

( A) ( x ) FAi ( y ),
i
Proposition 2.15. If F1 and F2 are two neutrosophic sets FNm
yx R
in U such that F1 F2, then N (F1) N (F2)
(i) N ( F1 F2 ) N ( F1 ) N ( F2 )
i
TNm ( x) TAi ( y ),
( A) yx R

(ii) N ( F1 F2 ) N ( F1 ) N ( F2 ) i
I Nm ( x) I Ai ( y ),
( A) yx R
Proposition 2.16. i
FNm ( x) FAi ( y )
( A) yx R
(i) N ( F ) ~ N (~ F )
such that,
(ii) N (~ F ) ~ N (~ F )
( A) ( x) , I Nm( A) ( x) , FNm( A) ( x) 0,1 ,
i i i
TNm
(iii) N ( F ) N ( F )
i
TNm ( A)
( x), I Nm
i
( A)
( x), FNm
i
( A)
( x)0,1
,

0 TNm
i
( A) ( x) I Nm( A) ( x) FNm( A) ( x) 3, and
i i

0 TNm
i
( A)
( x) I Nm
i
( A)
( x) FNm
i
( A)
( x) 3

Neutrosophic Sets and Systems, Vol. 16, 2017 85

Here and denote min and max operators re- Nm(B) = {p1, p4, p5}
spectively. TAi ( y) , I Ai ( y) and FAi ( y) are the membership se-
quences, indeterminacy sequences and non-membership ={<p1, (0.5, 0.6, 0.5), (0.3, 0.4, 0.3), (0.4, 0.3, 0.3)>,
sequences of y with respect to A and i =1, 2, , p. <p4, (0.5, 0.6, 0.5), (0.3, 0.4, 0.3), (0.4, 0.3, 0.3)>, <p5,
(0.2, 0.1, 0.0), (0.3, 0.2, 0.2), (0.8, 0.7, 0.7)>} and
Since Nm(A) and Nm ( A) are two neutrosophic multi- Nm(B) = {p1, p4, p5}
sets in U, thus neutrosophic multisets mappings
Nm , Nm : NmU NmU are respectively referred to as ={<p1, (0.5, 0.6, 0.5), (0.3, 0.4, 0.3), (0.4, 0.3, 0.3)>,
the upper and lower rough neutrosophic multisets approx- <p4, (0.5, 0.6, 0.5), (0.3, 0.4, 0.3), (0.4, 0.3, 0.3)>, <p5,
imation operators, and the pair is (Nm(A), Nm ( A)) called the (0.2, 0.1, 0.0), (0.3, 0.2, 0.2), (0.8, 0.7, 0.7)>}
rough neutrosophic multisets in (U, R).
Obviously, Nm(B) = Nm(B) is a definable neutrosophic
From the above definition, we can see that Nm(A) and multisets in the approximation space (U, R).
Nm ( A) have constant membership on the equivalence clas-
ses of U, if Nm(A) Nm(A) ; i.e., Definition 3.3 Let Nm( A) ( Nm( A), Nm( A)) be a rough
neutrosophic multisets in (U, R). The rough complement of
Nm(A) is denoted by ~ Nm( A) ( Nm( A)c , Nm( A)c ) where
( A) ( x) TNm( A) ( x),
i i
TNm
Nm( A)c and Nm( A)c are the complements of neutrosophic
( A) ( x) I Nm( A) ( x),
i i
I Nm multisets of Nm (A) and Nm(A) respectively,

## Nm( A) c = {x, ( FNm( A) ( x) , 1 I Nm( A) ( x) , TNm( A) ( x) )

i i i
( A) ( x) FNm( A) ( x).
i i
FNm
| x U},
Let consider the following example.
Nm( A) c = {x, ( FNm( A) ( x) , 1 I Nm( A) ( x ) , TNm( A) ( x ) )
i i i
Example 3.2 Let U = { p1, p2, p3, p4, p5, p6, p7, p8 } be
the universe of discourse. Let R be an equivalence relation | x U}
its partition of U is given by
U/R = {{p1, p4}, {p2, p3, p6}, { p5}, { p7, p8}}. where i = 1, 2, , p.
Example 3.4 Consider RNM, Nm(A) in the set X = {x1, x2,
Let Nm(A) = {(<p1, (0.8, 0.6, 0.5), (0.3, 0.2, 0.1), (0.4, 0.2,
x3, x4, x5, x6, x7}, y [x]R is equivalence relation and i = 1,
0.1)>, <p4, (0.5, 0.4, 0.3), (0.4, 0.4, 0.3), (0.6, 0.3, 0.3)>),
<p5, (0.2, 0.1, 0.0), (0.3, 0.2, 0.2), (0.8, 0.7, 0.7)>, <p7, (0.7, 2, 3.
0.6, 0.5), (0.3, 0.2, 0.1), (0.4, 0.3, 0.2)> } be a neutrosophic
multisets of U. By definition 3.1 we obtain: Let Nm(A) = {x1, [(0.6, 0.4, 0.4), (0.7, 0.3, 0.4)], [(0.8,
0.4, 0.5), (0.7, 0.6, 0.5)], [(0.4, 0.3, 0.5), (0.3, 0.2, 0.7)] ,
x2, [(0.4, 0.3, 0.3), (0.5, 0.3, 0.4)], [(0.2, 0.4, 0.4), (0.3,
Nm(A) = {p1, p4, p5} 0.3, 0.5)], [(0.7, 0.8, 0.4), (0.7, 0.1, 0.5)], x4, [(0.2, 0.5,
={<p1, (0.5, 0.6, 0.5), (0.3, 0.4, 0.3), (0.4, 0.3, 0.3)>, 0.7), (0.7, 0.8, 0.0)], [(1.0, 1.0, 0.0), (0.9, 0.2, 0.5)], [(0.1,
<p4, (0.5, 0.6, 0.5), (0.3, 0.4, 0.3), (0.4, 0.3, 0.3)>, <p5, 0.5, 0.3), (0.2, 0.8, 0.5)] }
(0.2, 0.1, 0.0), (0.3, 0.2, 0.2), (0.8, 0.7, 0.7)>} and Then the complement of Nm(A) is defined as
Nm( A) = {p1, p4, p5, p7, p8} ~ Nm( A) ( Nm( A)c , Nm ( A)c )
=
{<p1, (0.8, 0.4, 0.3), (0.4, 0.2, 0.1), (0.6, 0.2, 0.1)>, <p4, {x1, [(0.4, 0.6, 0.6), (0.4, 0.7, 0.7)], [(0.5, 0.6, 0.8),
(0.8, 0.4, 0.3), (0.4, 0.2, 0.1), (0.6, 0.2, 0.1)>, <p5, (0.2, (0.5, 0.4, 0.7)], [(0.5, 0.7, 0.4), (0.7, 0.8, 0.3)] , x2,
0.1, 0.0), (0.3, 0.2, 0.2), (0.8, 0.7, 0.7)>, <p7, (0.7, 0.6, [(0.3, 0.7, 0.4), (0.4, 0.7, 0.5)], [(0.4, 0.6, 0.2), (0.5,
0.5), (0.3, 0.2, 0.1), (0.4, 0.3, 0.2)>, <p8, (0.7, 0.6, 0.5), 0.7, 0.3)], [(0.4, 0.2, 0.7), (0.5, 0.9, 0.7)], x4, [(0.7,
(0.3, 0.2, 0.1), (0.4, 0.3, 0.2)>}. 0.5, 0.2), (0.0, 0.2, 0.7)], [(0.0, 0.0, 1.0), (0.5, 0.8,
0.9)], [(0.3, 0.5, 0.1), (0.5, 0.2, 0.2)] }.
For another neutrosophic multisets
Nm(B) = {<p1, (0.8, 0.6, 0.5), (0.3, 0.2, 0.1), (0.4, 0.2,
0.1)>, <p4, (0.5, 0.4, 0.3), (0.4, 0.4, 0.3), (0.6, 0.3, 0.3)>,
<p5, (0.2, 0.1, 0.0), (0.3, 0.2, 0.2), (0.8, 0.7, 0.7)>}.
The lower approximation and upper approximation of
Nm(B) are calculated as

86 Neutrosophic Sets and Systems, Vol. 16, 2017

Definition 3.5 Let Nm (A) and Nm(B) are RNM respec- [(0.3, 0.4, 0.5), (0.6, 0.4, 0.0)], [(1.0, 1.0, 0.0), (0.7,
tively in U, then the following definitions hold: 0.2, 0.5)], [(0.1, 0.5, 0.3), (0.2, 0.8, 0.5)], x4, [(0.2,
0.5, 0.7), (0.7, 0.8, 0.2)], [(1.0, 1.0, 0.0), (0.9, 0.2,
(i) Nm(A) = Nm(B) iff Nm ( A) Nm ( B) and
0.5)], [(0.1, 0.5, 0.3), (0.2, 0.8, 0.7)] }.
Nm ( A) Nm( B)
Proposition 3.7 If Nm, Mm, Lm are the RNM in (U, R),
(ii) Nm(A) Nm(B) iff Nm ( A) Nm ( B) and then the following propositions are stated from definitions.
Nm ( A) Nm( B) (i) (~Nm) = Nm

## (iii) ( Nm Mm) Lm Nm (Mm Lm), and

Nm ( A) Nm ( B), Nm ( A) Nm ( B )
( Nm Mm) Lm Nm (Mm Lm)
(iv) Nm ( A) Nm( B) (iv) ( Nm Mm) Lm ( Nm Mm) ( Nm Lm), and

## (v) Nm ( A) Nm( B) Proof (i):

(~ Nm( A)) ~ (~ ( Nm( A), Nm( A)))
Nm ( A) Nm ( B ), Nm ( A) Nm ( B )
~ ( Nm( A)c , Nm( A)c )
(vi) Nm ( A) Nm( B)
( Nm( A), Nm( A))
Nm ( A) Nm ( B), Nm ( A) Nm ( B)
= Nm(A)
Example 3.6 Consider Nm(A) in Example 3.4 and Nm(B) Proof (ii iv) : The proofs is straightforward from defini-
are two RNM. tion.
Nm(B) = {x1, [(0.6, 0.1, 0.2), (0.3, 0.3, 0.3)], [(0.7, 0.2, Proposition 3.8 De Morgans Law are satisfied for
0.5), (0.8, 0.6, 0.5)], [(0.7, 0.3, 0.5), (1.0, 0.2, 0.7)] , rough neutrosophic multisets:
x2, [(0.4, 0.4, 0.7), (0.6, 0.5, 0.6)], [(0.3, 0.4, 0.4),
(0.6, 0.2, 0.5)], [(0.7, 0.8, 0.4), (0.6, 0.1, 0.5)], x3, (i) (Nm(A) Nm(B)) = ( Nm(A)) ( ~ Nm(B))
[(0.3, 0.4, 0.5), (0.6, 0.4, 0.0)], [(1.0, 1.0, 0.0), (0.7,
(ii) (Nm(A) Nm(B)) =( Nm(A)) (~ Nm(B))
0.2, 0.5)], [(0.1, 0.5, 0.3), (0.2, 0.8, 0.5)] , x4, [(0.4,
0.5, 0.6), (0.7, 0.8, 0.2)], [(1.0, 1.0, 0.0), (0.9, 0.2, Proof (i):
0.1)], [(0.6, 0.5, 0.3), (0.2, 0.2, 0.7)] }
(Nm(A) Nm(B))
Then, we have
= ({ Nm(A) Nm(B) }, { Nm( A) Nm( B) })
(i) Nm(A) Nm(B)
(ii) Nm ( A) Nm( B) = (~{ Nm(A) Nm(B) }, ~{ Nm( A) Nm( B) })

= {x1, [(0.6, 0.1, 0.2), (0.7, 0.3, 0.3)], [(0.8, 0.2, = ({ Nm(A) Nm(B) }c, { Nm( A) Nm( B) }c)
0.5), (0.7, 0.6, 0.5)], [(0.7, 0.3, 0.5), (1.0, 0.2, 0.7)] ,
x2, [(0.4, 0.3, 0.3), (0.6, 0.3, 0.4)], [(0.3, 0.4, 0.4), = (~{ Nm(A) Nm(B) }, ~{ Nm( A) Nm( B) })
(0.6, 0.2, 0.5)], [(0.7, 0.8, 0.4), (0.7, 0.1, 0.5)], x3,
[(0.3, 0.4, 0.5), (0.6, 0.4, 0.0)], [(1.0, 1.0, 0.0), (0.7, = (Nm(A)) (~Nm(B)).
0.2, 0.5)], [(0.1, 0.5, 0.3), (0.2, 0.8, 0.5)], x4, [(0.4,
Proof (ii): Similar to the proof of (i).
0.5, 0.6), (0.7, 0.8, 0.0)], [(1.0, 1.0, 0.0), (0.9, 0.2,
0.1)], [(0.6, 0.5, 0.3), (0.2, 0.2, 0.5)] }. Proposition 3.9. If A and B are two neutrosophic multi-
sets in U such that A B, then Nm(A) Nm(B)
(iii) Nm ( A) Nm( B)
(i) Nm( A B) Nm( A) Nm( B)
= {x1, [(0.6, 0.4, 0.4), (0.3, 0.3, 0.4)], [(0.7, 0.4,
0.5), (0.7, 0.6, 0.5)], [(0.4, 0.3, 0.5), (0.3, 0.2, 0.7)], (ii) Nm( A B) Nm( A) Nm( B)
x2, [(0.4, 0.4, 0.7), (0.5, 0.5, 0.6)], [(0.2, 0.4, 0.4),
(0.3, 0.3, 0.5)], [(0.7, 0.8, 0.4), (0.6, 0.1, 0.5)], x3,

Neutrosophic Sets and Systems, Vol. 16, 2017 87

## Proof (i): where

( A B) ( x) inf {TNm( A B) ( x) | x X } ( x) T Ai ( y ),
i i i
TNm TNm (~ A) yx R

## inf ( max {TNm

i
( A) ( x),TNm( B ) ( x) | x X })
i i
I Nm (~ A)
( x) I Ai ( y ),
yx R

max{inf {TNm
i
( A) ( x) | x X }, inf {TNm( B) ( x) | x X }}
i i
FNm (~ A)
( x) FAi ( y ),
yx R
max{(TNm
i
( A) ( x) , TNm( B) ( x))| x X }
i
Hence Nm(A) = Nm(~ A) .
(TNm
i
( A) TNm( B) ) ( x)
i
Proof (ii): Similar to the proof of (i).
Similarly, Proof (iii): For any y Nm(A), we can have

( A B) ( x) ( I Nm( A) I Nm( B) ) ( x) , ( A) ( x) TA ( y) TA ( y) ,
i i i i i i
I Nm TNm
yx R yx R

( A B) ( x) ( FNm( A) FNm( B) ) ( x)
i i i
FNm
( A) ( x) I A ( y ) I A ( y) , and
i i i
I Nm
yx R yx R
Thus, Nm( A B) Nm( A) Nm( B)
( A) ( x) FA ( y) FA ( y)
i i i
FNm
Hence, yx R yx R

## Nm( A B) Nm( A) Nm( B) Hence Nm( A) Nm( A) .

Proof (ii): Similar to the proof of (i).
Proposition 3.10. Conclusion
(i) Nm(A) = Nm(~ A) This paper firstly defined the rough neutrosophic mul-
tisets (RNM) theory and their properties and operations
were studied. The RNM are the extension of rough neutro-
(ii) Nm( A) ~ Nm (~ A)
sophic sets . The future work will cover the others op-
eration in rough set, neutrosophic multisets and rough neu-
(iii) Nm( A) Nm( A) trosophic set that is suitable for RNM theory such as the
Proof (i): According to Definition 3.1, we can obtain notion of inverse, symmetry, and relation.

## A = {x, ( TAi ( x), I Ai ( x), FAi ( x) ) | x X} References

 L. A. Zadeh, Fuzzy sets, Inf. Control, vol. 8(3) (1965),
~A = {x, ( FAi ( x),1 I Ai ( x),TAi ( x) ) | x X}
pp. 338353.
 K. T. Atanassov, Intuitionistic Fuzzy Set, Fuzzy Sets
Nm(~ A) = {x, ( FNm
i
(~ A)
( x), 1- I Nm
i
(~ A)
( x),
and Systems, vol. 20 (1986), pp. 8796.
 Z. Pawlak, Rough sets, Int. J. Comput. Inf. Sci., vol.
i
TNm(~ A)
( x) ) | y [x]R, x U} 11(5) (1982), pp. 341356.
 P. K. Maji, R. Biswas, and A. R. Roy, Soft set theory,
~ Nm(~ A) = {x, ( TNm
i
(~ A)
( x), 1- (1- I Nm
i
(~ A)
( x)), Comput. Math. with Appl., vol. 45(4-5) (2003), pp. 555
562.
i
FNm(~ A)
( x) ) | y [x]R, x U}  A. Mukherjee, Generalized Rough Sets: Hybrid Structure
and Applications. Springer, India, 2015.
= {x, ( TNm
i
( x), I Nm
i
( x),  Lirong Jian, S. Liu, and Y. Lin, Hybrid Rough Sets and
(~ A) (~ A)
Applications in Uncertain Decision-Making. CRC Press,
Taylor & Francis Group, New York, 2011.
i
FNm(~ A)
( x) ) | y [x]R, x U}
 F. Smarandache, A Unifying Field in Logics:
Neutrosophic Logic, Neutrosophy, Neutrosophic Set,
Neutrosophic Probability and Statistic, American
Research Press, 4th Edition, 2005.

88 Neutrosophic Sets and Systems, Vol. 16, 2017

 B. Said, M. Talea, A. Bakali, and F. Smarandache,  K. Mondal, S. Pramanik, and F. Smarandache, Several
Single Valued Neutrosophic Graphs, in New Trends in Trigonometric Hamming Similarity Measures of Rough
Neutrosophic Theory and Applications, 2016, pp 187-202. Neutrosophic Sets and their Applications in Decision, in
 I. Kandasamy and F. Smarandache, Triple Refined New Trends in Neutrosophic Theory and Applications,
Indeterminate Neutrosophic Sets for Personality 2016, pp. 93-103.
Classification Triple Refined Indeterminate Neutrosophic  I. Deli, S. Broumi, and F. Smarandache, On Neutrosophic
Sets for Personality Classification, in New Trends in Refined Sets and Their Applications in Medical. J. of
Neutrosophic Theory and Applications , 2016. New Theory, (6) (2015), pp. 8898.
 M. ahin, S. Alkhazaleh, and V. Uluay, Neutrosophic  F. Smarandache, Neutrosophic Set - a generalization of
Soft Expert Sets, Appl. Math., vol. 6 (2015), pp. 116 Intuitionistics Fuzzy Sets, Int. J. of Pure and Applied
127. Math, vol. 24 (3) (2005), pp. 287297.
 I. Deli, Interval-valued neutrosophic soft sets and its  F. Smarandache, Neutrosophic SetA Generalization of
decision making, Int. J. Mach. Learn. Cybern., (2015). the Intuitionistic Fuzzy Set, J. Def. Resour. Manag.,(1)
 M. Ali and F. Smarandache, Complex neutrosophic set, (2010), pp. 107116.
Neural Comp. and Applications, (2016), pp. 118.  S. Broumi, I. Deli, and F. Smarandache, Neutrosophic
 M. Abdel-baset, I. M. Hezam, and F. Smarandache, soft multiset theory, Ital. J. Pure Appl. Math., vol. 32
Neutrosophic Goal Programming, Neutrosophic Sets (2014), pp. 503514.
and System, vol. 11(2016) pp. 112118.  S. Broumi, I. Deli, and F. Smarandache, Relations on
 W. B. V. Kandasamy and F. Smarandache, Neutrosophic neutrosophic multi sets with properties, (2015), pp. 115.
Rings. Hexis Phoenix, Arizona, 2006.  S. Broumi, I. Deli, and F. Smarandache, Neutrosophic
 S. Broumi, F. Smarandache, and M. Dhar, Rough Multi relations and Their Properties, (2014), pp. 118.
Neutrosophic Sets, Ital. J. Pure Appl. Math., vol. 32  M. Sa, R. Chatterjee, P. Majumdar, and S. K. Samanta,
(2014), pp. 493502. Single valued neutrosophic multisets, Annals of Fuzzy
 S. Broumi and F. Smarandache, Interval Neutrosophic Mathematics and Informatics, vol. x (x) (2015), pp. 499
Rough Set, Neutrosophic Sets and Systems, vol. 7 (2014), 514.
pp. 2331.
 S. Pramanik and K. Mondal, Cosine Similarity Measure
Of Rough Neutrosophic Sets And Its Application In
Medical Diagnosis, Global J. and Adv. Research, vol. 2
(1) (2015), pp. 212220.
 K. Mondal and S. Pramanik, Rough Neutrosophic Multi-
Attribute Decision-Making Based on Grey Relational
Analysis, Neutrosophic Sets and System., vol. 7 (2015),
pp. 817. Received: June 6, 2017. Accepted: June 21, 2017.