ISSN: 1304-7981 Number: 4, Year: 2014, Pages: 33-41. http://jnrs.gop.edu.

tr

Received : 21.10.2012 Editors-in-Chief : Hikmet Günal

Accepted : 12.01.2014 Area Editor : Oktay Muhtaroğlu

Contributions to the Theory of Soft Sets

Naim Çağman ([email protected])

Department of Mathematics, Gaziosmanpaşa University, 60250 Tokat, Turkey

Abstract - Molodtsov introduced the concept of soft set as

a new mathematical tool for dealing with problems containing
uncertainties. In the literature, different kinds of operations of
soft sets are defined and used in theory and applications. Some
researchers showed that some of these definitions have a few gaps. Keywords - Soft sets, soft in-
In this paper, we make contributions to the theory of soft sets
tersection, soft union, soft com-
in which we first make some modifications to the operations of
soft sets and fill in these gaps. Until now, to define soft sets plement, soft products.
and their properties, the authors have used different subsets of
the parameters set for each soft set. Whereas, we use a single
parameter set for all of soft sets in our approach. We then compare
the definitions with those defined before.

1 Introduction
Problems with uncertainties are a major issue in many fields such as economics, engi-
neering, environment and so on. One of the theories that deal with uncertainties is the
soft set theory which is introduced by Molodtsov [4] as a new mathematical tool this.
In recent years, works on soft set theory and its applications have been progressing
rapidly. In the literature, different kinds of operations of soft sets are defined and used
in the works on soft set theory and its applications. Well known operations of soft sets
and their properties are given by Maji et al. [3]. But some of these definitions and their
properties have a few gaps, which have been pointed out by Ali et al. [1] and Yang [9].
To make some modifications to the operations of soft sets and fill in these gaps, Ali
et al. [1], Cagman and Enginoglu [2], Neog and Sut [5], Pai and Miao [6], Sezgin and
Atagun [7], Xia and Zuhua [8], and Zhu and Wen [10] make contributions.
A soft set is a mapping of parameters set into the set of all subsets of the universe.
Until now, to define soft sets, the authors have used different subsets of the parameters
set for each soft set. This has led to the emergence of some problems in the operations
and properties of soft sets. The aim of this paper is to make contributions to the
Journal of New Results in Science 4 (2014) 33-41 34

theory of soft sets, in which we first give a new approach to the soft sets by using a
single parameter set for all of soft sets. This new approach makes the operations and
their properties clear and free of the difficulties mentioned above. We then compare
the definitions with other recently defined operations.

2 Definition of Soft Sets

Let U be an initial universe set and let E be a set of parameters. Then, to use an
adequate parametrization, Molodtsov [4] gave the definition of soft sets as follows;

Definition 2.1. [4] A pair (F, E) is called a soft set (over U ) if and only if F is a
mapping of E into the set of all subsets of the set U .

Where F is called approximate function and E is called parameters set of the soft
set (F, E). Each parameter x ∈ E may be a fuzzy word, fuzzy sentence, real number
and so on. For x ∈ E, the set F (x) is called x-approximation of the soft set (F, E)
which may be arbitrary, some of them may be empty and some may have a nonempty
intersection.
From now on, the soft set (F, E) will be shortly denoted by FE .
A soft set FA over U can be written as a set of ordered pairs,

FE = {(x, F (x)) : x ∈ E}

Elements of a soft set with empty approximation are normally not listed.
A soft set is a mapping of parameters set into the set of all subsets of the universe.
Until now, to define the soft sets and their operations, the authors have used different
subsets of the parameters set for each soft set. To illustrate, let us consider following
convenient example, which is an expansion of Molodtsov’s example given in [4].

Example 2.2. Assume that there are six houses in the universe U = {hl , h2 , h3 , h4 , h5 , h6 }
and E = {e1 , e2 , e3 , e4 , e5 } is the set of parameters. Here, ei (i = 1, 2, 3, 4, 5) stand for
the parameters “expensive”, “beautiful”, “wooden”, “cheap”, and “in green surround-
ings” respectively. In this case, to define a soft set means to point out expensive houses,
beautiful houses, and so on. Then, following soft sets describe the attractiveness of the
houses which a couple is going to buy.
Suppose that according to Mr. X’s, F (e1 ) = {h2 , h3 , h4 }, F (e2 ) = U , F (e3 ) = U ,
F (e4 ) = ∅, and F (e5 ) = {h1 , h2 , h5 , h6 }. If Mr. X considers a subset A = {e1 , e4 , e5 }
or B = {e2 , e3 , e4 , e5 } of E, then soft set FA or FB can be written by

FA = {(e1 , {h2 , h3 , h4 }), (e5 , {h1 , h2 , h5 , h6 })} or

FB = {(e2 , U ), (e3 , U ), (e5 , {h1 , h2 , h5 , h6 })}

Suppose that according to Mrs. X’s, G(e1 ) = ∅, G(e2 ) = {h3 , h6 }, G(e3 ) = U ,

G(e4 ) = ∅, and G(e5 ) = {h2 , h4 , h6 }. If Mrs. X considers a subset A = {e1 , e4 , e5 } or
B = {e2 , e3 , e4 , e5 } of E, then soft set GA or GB can be written by

GA = {(e5 , {h2 , h4 , h6 })} or

GB = {(e2 , {h3 , h6 }), (e3 , U ), (e5 , {h2 , h4 , h6 })}
Journal of New Results in Science 4 (2014) 33-41 35

3 Operations of Soft Sets

In this section, we give contributions to the theory of soft sets in which we will use only
a single parameter set for all soft sets. This new approach make the operations and
their properties clear and free of the difficulties mentioned in the first section.
From now on, the single parameters set to be used is denoted by E to define the soft
sets and their operations and therefore the subscript E can be deleted from the soft
sets FE , i.e, a soft set FE will be denoted shortly by F unless this will cause confusion.
In general, soft sets and their approximate functions are denoted by F, G, H, ..., and
soft sets and their approximate functions are used interchangeably.

Example 3.1. Let us consider Example 2.2. Then, two soft sets F and G describe the
attractiveness of the houses which a couple is going to buy.
Mr. X’s soft set F can be written by

F = {(e1 , {h2 , h3 , h4 }), (e2 , U ), (e3 , U ), (e5 , {h1 , h2 , h5 , h6 })}

Mrs. Y’s soft set G can be written by

G = {(e2 , {h3 , h6 }), (e3 , U ), (e5 , {h2 , h4 , h6 })}

Definition 3.2. A soft sets F over U is said to be a null soft set, denoted by Φ, if
F (x) = ∅ for all x ∈ E.

Definition 3.3. A soft sets F over U is said to be absolute soft set, denoted by U, if
F (x) = U for all x ∈ E.

Remark 3.4. Definitions 3.2 and 3.3 were first given by Maji et al. ([3], Definition
2.7 and 2.8) which are inadequate according to the definition of soft sets ([3], Definition
2.1). For two different soft sets FA and FB over U have own absolute and null soft sets
which may be different. It means that there are more than one absolute and null soft
sets in the soft set theory where the definitions can be given with different subsets of
parameters set.
For example, let us consider Example 2.2. Then,
If A = {e3 }, then FA = à and GA = Ã, where à is the absolute soft set.
If B = {e2 , e3 }, then FB = B̃ and GA 6= B̃, where B̃ is the absolute soft set.
If C = {e4 }, then FC = ΦC and GC = ΦC , where ΦC is the null soft set.
If D = {e1 , e4 }, then FD 6= ΦC and GD = ΦD , where ΦD is the null soft set.
In this case, what can we say about Ãc̃ =?, B̃ c̃ =?, Φc̃C =? and Φc̃D =?.
˜
Definition 3.5. Let F, G ∈ S. Then, F is a soft subset of G, denoted by F ⊆G, if
F (x) ⊆ G(x) for all x ∈ E.

Proposition 3.6. If F ∈ S, then

˜
1. Φ⊆F
˜
2. F ⊆F
˜
3. F ⊆U
Journal of New Results in Science 4 (2014) 33-41 36

Definition 3.7. Let F, G ∈ S. Then, F equal to G, denoted by F = G, if F (x) = G(x)

for all x ∈ E.
Proposition 3.8. If F, G, H ∈ S, then
˜ G⊆F
1. F ⊆G, ˜ ⇔F =G

˜ G⊆H
2. F ⊆G, ˜ ⇒ F ⊆H
˜

˜ G, is a
Definition 3.9. Let F, G ∈ S. Then, soft union of F and G, denoted by F ∪
soft set over U whose approximate function is defined by

˜ G : E → P (U ),
F∪ ˜ G)(x) = F (x) ∪ G(x)
(F ∪

Proposition 3.10. If F ∈ S, then

˜F = F
1. F ∪
˜Φ = F
2. F ∪
˜U = U
3. F ∪
˜ G,
Definition 3.11. Let F, G ∈ S. Then, soft intersection of F and G, denoted by F ∩
is a soft set over U whose approximate function is defined by

˜ G : E → P (U ),
F∩ ˜ G)(x) = F (x) ∩ G(x)
(F ∩

Proposition 3.12. If F ∈ S, then

˜F = F
1. F ∩
˜Φ = Φ
2. F ∩
˜U = F
3. F ∩
Proposition 3.13. If F, G ∈ S, then
˜ G = G∩
1. F ∩ ˜F

˜ G = G∪
2. F ∪ ˜F

Proposition 3.14. If F, G, H ∈ S, then

˜ (G∩
1. F ∩ ˜ H) = (F ∩
˜ G)∩
˜H

˜ (G∪
2. F ∪ ˜ H) = (F ∪
˜ G)∪
˜H

˜ (G∪
3. F ∩ ˜ H) = (F ∩
˜ G)∪
˜ (F ∩
˜ H)

˜ (G∩
4. F ∪ ˜ H) = (F ∪
˜ G)∩
˜ (F ∪
˜ H)
˜ is
Definition 3.15. Let F, G ∈ S. Then, soft difference of F and G, denoted by F \G,
a soft set over U whose approximate function is defined by
˜ : E → P (U ),
F \G ˜
(F \G)(x) = F (x) \ G(x)
Journal of New Results in Science 4 (2014) 33-41 37

Proposition 3.16. If F ∈ S, then

˜ =Φ
1. F \F
˜ =F
2. F \Φ
˜ =Φ
3. Φ\F
˜ =Φ
4. F \U
Definition 3.17. Let F ∈ S. Then, soft complement of F , denoted by F c̃ , is a soft set
over U whose approximate function is defined by
F c̃ : E → P (U ), F c̃ (x) = U \ F (x)
Note that to keep clear of confusion, for the soft sets operations we use classical sets
operations with tilde.
Proposition 3.18. If F ∈ S, then
1. (F c̃ )c̃ = F
2. Φc̃ = U
˜ F c̃ = U
3. F ∪
˜ F c̃ = Φ
4. F ∩
˜
˜ Gc̃ = F \G.
Proposition 3.19. If F, G ∈ S, then F ∩
Proof: The proofs can be made by using the approximate functions. For all x ∈ E,
˜ Gc̃ )(x) =
(F ∩ F (x) ∩ Gc̃ (x)
= F (x) ∩ (U \ G(x))
= F (x) ∩ (U ∩ (G(x))c )
= F (x) ∩ (G(x))c
= F (x) \ G(x)
= ˜
(F \G)(x)
Proposition 3.20. If F, G ∈ S, then the following De Morgan’s types of results are
true.
˜ G)c̃ = F c̃ ∩
1. (F ∪ ˜ Gc̃
˜ G)c̃ = F c̃ ∪
2. (F ∩ ˜ Gc̃
Proof: Let us prove first one, the other can be made similarly. For all x ∈ E,
˜ G)c̃ (x) =
(F ∪ U \ (F ∪ ˜ G)(x)
= [(F ∪ ˜ G)(x)]c
= [F (x) ∪ G(x)]c
= (F (x))c ∩ (G(x))c
= (U \ F (x)) ∩ (U \ G(x))
= F c̃ (x) ∩ Gc̃ (x)
= ˜ Gc̃ )(x)
(F c̃ ∩
Journal of New Results in Science 4 (2014) 33-41 38

Example 3.21. Let us consider Example 3.1 where two soft sets F and G are given
as;
F = {(e1 , {h2 , h3 , h4 }), (e2 , U ), (e3 , U ), (e5 , {h1 , h2 , h5 , h6 })}
G = {(e2 , {h3 , h6 }), (e3 , U ), (e5 , {h2 , h4 , h6 })}
Then, we can find the following;
F∩ ˜ G = {(e2 , {h3 , h6 }), (e3 , U ), (e5 , {h2 , h6 })}
F∪ ˜ G = {(e1 , {h2 , h3 , h4 }), (e2 , U ), (e3 , U ), (e5 , {h1 , h2 , h4 , h5 , h6 })}
˜
F \G = {(e1 , {h2 , h3 , h4 }), (e2 , {h1 , h2 , h4 , h5 }), (e5 , {h1 , h5 })}
˜
G\F = {(e5 , {h4 })}
F c̃ = {(e1 , {h1 , h5 , h6 }), (e4 , U }), (e5 , {h3 , h4 })}
Gc̃ = {(e1 , U ), (e2 , {h1 , h2 , h4 , h5 }), (e4 , U }), (e5 , {h1 , h3 , h5 })}
˜ Gc̃ = {(e1 , {h1 , h5 , h6 }), (e4 , U }), (e5 , {h3 })}
F c̃ ∩
˜ Gc̃ = {(e1 , U ), (e2 , {h1 , h2 , h4 , h5 }), (e4 , U }), (e5 , {h1 , h3 , h4 , h5 })}
F c̃ ∪
˜ G)c̃ = {(e1 , U ), (e2 , {h1 , h2 , h4 , h5 }), (e4 , U }), (e5 , {h1 , h3 , h4 , h5 })}
(F ∩
˜ G)c̃ = {(e1 , {h1 , h5 , h6 }), (e4 , U }), (e5 , {h3 })}
(F ∪
Until now, we defined the binary operations of soft sets which depend on an ap-
proximate function of one variable. Now, we define products of the soft sets which are
binary operations of soft sets depending on an approximate function of two variables.
Definition 3.22. Let F, G ∈ S. Then, AN D-product of two soft sets F and G, denoted
by F ∧ G, is a binary operation of the soft sets over U whose approximate function is
defined by
F ∧ G : E × E → P (U ), (F ∧ G)(x, y) = F (x) ∩ G(y)
Definition 3.23. Let F, G ∈ S. Then, OR-product of two soft sets F and G, denoted
by F ∨ G, is a binary operation of the soft sets over U whose approximate function is
defined by
F ∨ G : E × E → P (U ), (F ∨ G)(x, y) = F (x) ∪ G(y)
Note that ∧ and ∨ are not commutative.
Proposition 3.24. If F, G, H ∈ S, then
1. F ∨ (G ∨ FC ) = (F ∨ G) ∨ H
2. F ∧ (G ∧ H) = (F ∧ G) ∧ H
Proposition 3.25. If F, G, H ∈ S, then the following De Morgan’s types of results are
true.
1. (F ∨ G)c̃ = F c̃ ∧ Gc̃
2. (F ∧ G)c̃ = F c̃ ∨ Gc̃
Proof: Let us prove first one, the other can be made similarly. For all x, y ∈ E,
(F ∨ G)c̃ (x, y) = U \ (F ∨ G)(x, y)
= [(F ∨ G)(x, y)]c
= [F (x) ∪ G(y)]c
= (F (x))c ∩ (G(y))c
= (U \ F (x)) ∩ (U \ G(y))
= F c̃ (x) ∩ Gc̃ (y)
= (F c̃ ∧ Gc̃ )(x, y)
Journal of New Results in Science 4 (2014) 33-41 39

4 Comparision of the Definitions

In this section, we compare our definitions with the definitions given by Cagman and
Enginoglu [2] and Maji et al. [3].
Let us first compare the definitions of soft sets in Table 1.
In this paper, In [2], In [3],
Our approach by Cagman and Enginoglu by Maji et al.

F = {(x, F (x)) : x ∈ E} FA = {(x, F (x)) : x ∈ E} FA = {(x, F (x)) : x ∈ A}

where where where

E is a universe of parameters A⊆E A⊆E

and and and

½
A → P (U )
F : E → P (U ) F : F : A → P (U )
Ac → {∅}

Table 1: Definition of the soft sets

We can compere the union of soft sets as in Table 2.

In this paper In [2] In [3]

˜G = H
F∪ ˜ GB = HA∪B
FA ∪ ˜ GB = HA∪B
FA ∪
½
A ∪ B → P (U )
H : E → P (U ) H: H : A ∪ B → P (U )
(A ∪ B)c → {∅}

 F (x), if x ∈ A − B,
H(x) = F (x) ∪ G(x) H(x) = F (x) ∪ G(x) H(x) = G(x), if x ∈ B − A,

F (x) ∪ G(x), if x ∈ A ∩ B.

Table 2: Union of the soft sets

We can compere the intersection of soft sets as in the in Table 3.

In this paper In [2] In [3]

˜G = H
F∩ ˜ GB = HA∩B
FA ∩ ˜ GB = HA∩B
FA ∩
½
A ∩ B → P (U )
H : E → P (U ) H: H : A ∩ B → P (U )
(A ∩ B)c → {∅}

H(x) = F (x) ∩ G(x) H(x) = F (x) ∩ G(x) H(x) = F (x) or G(x)

Table 3: Intersection of the soft sets

We can compere the complement of soft sets as in Table 4.

Journal of New Results in Science 4 (2014) 33-41 40

In this paper In [2] In [3]

F c̃ is complement of F (FA )c̃ = FAc̃ (FA )c̃ = FeA

c̃

½
c̃ c̃ A → P (U )
F : E → P (U ) F : F c̃ :eA → P (U )
Ac → {∅}

F c̃ (x) = U − F (x) F c̃ (x) = U − F (x) F c̃ (x) = U − F (¬x)

Table 4: Complement of the soft sets

Where ¬x is not x, ¬(¬x) = x and eA = {¬x : x ∈ A}.

We can compere the AN D-product of soft sets as in the in Table 5.

In this paper In [2] In [3]

F ∧G=H FA ∧ GB = HA∧B FA ∧ GB = HA∧B

½
A × B → P (U )
H : E × E → P (U ) H: H : A × B → P (U )
(A × B)c → {∅}

H(x, y) = F (x) ∩ G(y) H(x, y) = F (x) ∩ G(y) H(x, y) = F (x) ∩ G(y)

Table 5: AN D-product of the soft sets

We can compere the OR-product of soft sets as in the in Table 6.

In this paper In [2] In [3]

F ∨G=H FA ∨ GB = HA∨B FA ∨ GB = HA∨B

½
A × B → P (U )
H : E × E → P (U ) H: H : A × B → P (U )
(A × B)c → {∅}

H(x, y) = F (x) ∪ G(y) H(x, y) = F (x) ∪ G(y) H(x, y) = F (x) ∪ G(y)

Table 6: OR-product of the soft sets

5 Conclusion
To deal with uncertainties, the soft set theory has been applied to many fields from
theoretical to practical. In the theory and applications, authors have defined and used
different kinds of operations of soft sets, some of which have a few gaps. Most of them
have been pointed out by Ali et al. [1]. They also have made some modifications to
the operations of soft sets to fill in these gaps under some conditions. In addition, they
have given definition of extended intersection, restricted intersection, restricted union,
restricted difference, relative null soft set, relative whole soft set, relative complement,
and have given their properties under some restrictions.
In this paper, we make contributions to the theory of soft sets, in which we use only
a single parameter set for all soft sets. This new approach makes operations and their
Journal of New Results in Science 4 (2014) 33-41 41

properties clear and free of the difficulties without conditions. We then compare the
definitions with other recently defined operations. We are confident that this work will
be beneficial for future studies on soft set theory.

References
[1] M.I. Ali, F. Feng, X. Liu, W.K. Min and M. Shabir, On some new operations in soft
set theory, Computers and Mathematics with Applications 57 (2009) 1547-1553.

[2] N. C.ağman and S. Enginoğlu, Soft set theory and uni-int decision making, Euro-
pean Journal of Operational Research 207/2 (2010) 848-855.

[3] P.K. Maji, R. Biswas and A.R. Roy, Soft set theory, Computers and Mathematics
with Applications 45 (2003) 555-562.

[4] D.A. Molodtsov, Soft set theory-first results, Computers and Mathematics with
Applications 37 (1999) 19-31.

[5] T. J. Neog, D.M. Sut, A New Approach to the Theory of Soft Sets, International
Journal of Computer Applications 32/2 (2011) 0975-8887.

[6] D. Pei, D. Miao, From soft sets to information systems, in: X. Hu, Q. Liu, A.
Skowron, T.Y. Lin, R.R. Yager, B. Zhang (Eds.), Proceedings of Granular Com-
puting, vol. 2, pp. 617-621, 2005, IEEE.

[7] A. Sezgin, A. O. Atagun, On operations of soft sets, Computers and Mathematics

with Applications 61 (2011) 1457-1467.

[8] Y. Xia, L. Zuhua, Some new operations of soft sets, Uncertainty Reasoning and
Knowledge Engineering (URKE), 2nd International Conference, pp. 290 - 295,
14-15 Aug. 2012, IEEE.

[9] C.F. Yang, A note on ”Soft set theory” [Computers and Mathematics with Ap-
plications 45 (2003) 555-562], Computers and Mathematics with Applications 56
(2008) 1899-1900.

[10] P. Zhu, Q. Wen, Operations on Soft Sets Revisited, Journal of Applied Mathemat-
ics 2013 (2013) 1-7.