Closed Sets in Generalized Topological Spaces: S. Maragathavalli, M. Sheik John, D. Sivaraj

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Journal of Advanced Research in Vol. 2, Issue. 1, 2010, pp.

57-64
Pure Mathematics doi: 10.5373/jarpm.243.102109
Online ISSN: 1943-2380

On g−closed sets in generalized topological spaces


S. Maragathavalli1 , M. Sheik John1 , D. Sivaraj2,∗
1
Department of Mathematics, N.G.M College, Pollachi, Tamil Nadu, India.
2
Department of Computer Science, D.J. Academy for Mangerial Excellence,
Coimbatore, Tamil Nadu, India.

Abstract. We define gµ −closed sets in generalized topological spaces and Igµ −closed
sets in generalized topological spaces with a hereditary class and give characterizations
and properties of these sets.

Keywords: µ−closed and µ−open sets; Generalized topology; Hereditary class; Quasi-topology.

Mathematics Subject Classification 2010: 54A05, 54A10.

1 Introduction
A generalized topology or simply GT [1] µ on a nonempty set X is a collection of
subsets of X such that ∅ ∈ µ and µ is closed under arbitrary union. Elements of µ are
called µ−open sets. A subset of A of X is said to be µ−closed if X − A is µ−open. The
pair (X, µ) is called a generalized topological space (GTS). By a space (X, µ), we will
always mean a GTS (X, µ). A space (X, µ) is said to be a quasi-topological space [4],
if µ is closed under finite intersection. Clearly, every topological space as well as a
quasi-topological space is a GTS. If A is a subset of a space (X, µ), then cµ (A) is the
smallest µ−closed set containing A and iµ (A) is the largest µ−open set contained in A.
A subset A of a space (X, µ) is said to be α−open [3] (resp., semiopen [3], preopen [3],
b−open [8], β−open [3] ) if A ⊂ iµ cµ iµ (A) (resp., A ⊂ cµ iµ (A), A ⊂ iµ cµ (A), A ⊂
iµ cµ (A) ∪ cµ iµ (A), A ⊂ cµ iµ cµ (A)). We will denote the family of all α−open sets by α,
the family of all semiopen sets by σ, the family of all preopen sets by π, the family of
all b−open sets by b and the family of all β−open sets by β. If κ ∈ {µ, α, σ, π, b, β}
and A is a subset of a space (X, κ), then cκ (A) is the smallest κ−closed set containing
A and iκ (A) is the largest κ−open set contained in A. Note that the operator cκ is


Correspondence to: D. Sivaraj, Department of Computer Science, D.J. Academy for Mangerial Excel-
lence, Coimbatore- 641 032, Tamil Nadu, India. Email: ttn− sivaraj@yahoo.co.in

Received: 21 October 2009, revised: 26 November 2009, accepted: 10 December 2009.
http://www.i-asr.com/Journals/jarpm/ 57 ⃝2010
c Institute of Advanced Scientific Research
58 On g−closed sets in generalized topological spaces

monotonic, increasing and idempotent and the operator iκ is monotonic, decreasing and
idempotent. Clearly, A is κ−open if and only if A = iκ (A) and A is κ−closed if and only
if A = cκ (A). Also, for every subset of A of a space (X, κ), X − iκ (A) = cκ (X − A). In
section 2, we define a new class of sets, called gκ −closed sets and give its characterizations
and properties. If κ = τ, a topology on X, then gκ −closed sets are nothing but g−closed
[7] sets in topological spaces. In section 3, we define another new class of subsets of
a space (X, κ) which contains properly the family of all gκ −closed sets and give its
characterizations and properties.

2 gκ −closed sets
Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. A subset A of X is said to be
gκ −closed if cκ (A) ⊂ M whenever A ⊂ M and M ∈ κ. Clearly, every κ−closed set is
gκ −closed and cκ (A) is gκ −closed for every subset A of X. The following Example 2.1
shows that the converse is not be true for κ = µ. Theorem 2.1 gives a characterization of
gκ −closed sets. Theorem 2.2 shows that a gκ −closed set A is a κ−closed set, if cκ (A)−A
is κ−closed.
Example 2.1. Let X = {a, b, c} and µ = {∅, {a} }. Since there is no µ−open set
containing {b}, {b} is gµ −closed but not µ−closed.
Theorem 2.1. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. Then a subset A
of X is gκ −closed if and only if F ⊂ cκ (A) − A and F is κ−closed imply that F = ∅.
Proof. Let F be a κ−closed subset of cκ (A) − A. Since A ⊂ X − F and A is gκ −closed,
cκ (A) ⊂ X − F and so F ⊂ X − cκ (A). Therefore, F = ∅. Conversely, suppose the
condition holds and A ⊂ M and M ∈ κ. If cκ (A) ̸⊂ M, then cκ (A) ∩ (X − M ) is a
nonempty κ−closed subset of cκ (A) − A, a contradiction to the hypothesis. Therefore,
cκ (A) ⊂ M which implies that A is gκ −closed.

Theorem 2.2. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. Then a gκ −closed
subset A of X is a κ−closed set if cκ (A) − A is a κ−closed set.
Proof. By Theorem 2.1, cκ (A) − A = ∅ and so cκ (A) = A which implies that A is
κ−closed.

The proof of the following Lemma 2.3 follows from Theorem 2.2 of [6]. Theo-
rem 2.4 below shows that in a quasi-topological space (X, µ), the union of two gµ −closed
sets is again a gµ −closed set. Example 2.2 shows that the condition quasi-topology on
the space cannot be replaced by generalized topology. Example 2.3 below shows that
the intersection of two gµ −closed sets need not be a gµ −closed set even in a topological
space. Theorem 2.5 shows that, the intersection of a gκ −closed set with a κ−closed is a
gκ −closed set where κ ∈ {µ, α, σ, π, b, β}.
Lemma 2.3. Let (X, µ) be a quasi-topological space. Then cµ (A ∪ B) = cµ (A) ∪ cµ (B)
for every subsets A and B of X.
S. Maragathavalli, M. Sheik John and D. Sivaraj 59

Theorem 2.4. Let (X, µ) be a quasi-topological space. Then the following hold.
(a) If A and B are gµ −closed subsets of X, then A ∪ B is also a gµ −closed set.
(b) If A and B are gα −closed subsets of X, then A ∪ B is also a gα −closed set.

Proof. (a) Suppose A and B are gµ −closed. Let M ∈ µ such that A ∪ B ⊂ M. Since A
and B are gµ −closed sets, cµ (A) ⊂ M and cµ (B) ⊂ M and so cµ (A) ∪ cµ (B) ⊂ M. By
Lemma 2.3, it follows that cµ (A ∪ B) ⊂ M and so the proof follows.
(b) The proof follows from (a) and Theorem 2.3 of [6], since (X, α) is a quasi-topological
space.

Example 2.2. Let X = {a, b, c} and µ = {∅, {b}, {c}, {a, b}, {a, c}, {b, c}, X}. Then µ
is a GT but not a quasi-topology. If A = {b} and B = {c}, then A and B are gµ −closed
sets but their union is not a gµ −closed set.

Example 2.3. Consider the topological space (X, τ ) where X = {a, b, c} with the topol-
ogy µ = {∅, {a}, X}. If A = {a, b} and B = {a, c}, then A and B are gµ −closed sets.
But A ∩ B = {a}, is not a gµ −closed set.

Theorem 2.5. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. If A is gκ −closed


and B is κ−closed, then A ∩ B is a gκ −closed set.

Proof. Suppose A ∩ B ⊂ M where M is κ−open. Then A ⊂ (M ∪ (X − B)). Since A is


gκ −closed, cκ (A) ⊂ (M ∪ (X − B)) and so cκ (A) ∩ B ⊂ M and so cκ (A ∩ B) ⊂ M which
implies that A ∩ B is a gκ −closed set.

The following Theorem 2.6 shows that every subset of a space (X, κ) is a gκ −closed
set if and only if the family of all κ−open sets coincides with the family of all κ−closed
sets. Theorem 2.7 below gives a property of gκ −closed sets.

Theorem 2.6. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. Then κ = {A |


X − A ∈ κ} if and only if every subset of X is a gκ −closed set.

Proof. Suppose κ = {A | X − A ∈ κ}, the family of all κ−closed sets. If A is a subset


of X such that A ⊂ M where M ∈ κ, then cκ (A) ⊂ cκ (M ) = M and so A is gκ −closed.
Conversely, suppose that every subset of X is a gκ −closed set. If M ∈ κ, since M is
gκ −closed, then cκ (M ) ⊂ M and so M is κ−closed and so κ ⊂ {A | X − A ∈ κ}. If
B ∈ {A | X − A ∈ κ}, then (X − B) ∈ κ and so cκ (X − B) ⊂ X − B. Therefore,
X − B is κ−closed and so B is κ−open. Hence {A | X − A ∈ κ} ⊂ κ which implies that
κ = {A | X − A ∈ κ}.

Theorem 2.7. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} and A and B be


subsets of X. If A ⊂ B ⊂ cκ (A) and A is gκ −closed, then B is gκ −closed.

Proof. If F is κ−closed such that F ⊂ cκ (B) − B, by hypothesis, F ⊂ cκ (A) − A. Since


A is gκ −closed, by Theorem 2.1, F = ∅ and so B is gκ −closed.
60 On g−closed sets in generalized topological spaces

A subset A of a space (X, κ) where κ ∈ {µ, α, σ, π, b, β} is said to be gκ −open


if its complement X −A is gκ −closed. Clearly, every κ−open set is gκ −open and iκ (A) is
gκ −open for every subset A of X. The following Theorem 2.8 follows from the definition
of gκ −closed sets. Since by Example 2.3, the intersection of two gµ −closed sets need
not be a gµ −closed set, the union of two gµ −open sets need not be a gµ −open set.
Theorem 2.9 below shows that the union of two gκ −open sets is a gκ −open set if they
are κ−separated. Two subsets A and B of a space (X, κ) are said to be κ−separated [2]
if A ∩ cκ (B) = cκ (A) ∩ B = ∅.

Theorem 2.8. A subset A of a space (X, κ) is gκ −open if and only if F ⊂ iκ (A)


whenever F is κ−closed and F ⊂ A.

Theorem 2.9. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. If A and B are
κ−separated, gκ −open sets, then A ∪ B is a gκ −open set.

Proof. Let F be a κ−closed set such that F ⊂ A ∪ B. Since A and B are κ−separated
sets, F ∩ cκ (A) ⊂ A and F ∩ cκ (B) ⊂ B. By Theorem 2.8, F ∩ cκ (A) ⊂ iκ (A) and
F ∩ cκ (B) ⊂ iκ (B) respectively. Now F = F ∩ (A ∪ B) = (F ∩ A) ∪ (F ∩ B) ⊂
(F ∩ cκ (A)) ∪ (F ∩ cκ (B)) ⊂ iκ (A) ∪ iκ (B) ⊂ iκ (A ∪ B). Therefore, again by Theorem 2.8,
A ∪ B is gκ −open.

The following Theorem 2.10 gives a characterization of gκ −open sets and The-
orem 2.11 below gives a property of gκ −open sets. Theorem 2.12 below gives a charac-
terization of gκ −closed sets in terms of gκ −open sets.

Theorem 2.10. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. A subset A of X


is gκ −open if and only if M = X whenever M is κ−open and iκ (A) ∪ (X − A) ⊂ M.

Proof. Suppose A is gκ −open and M is κ−open such that iκ (A) ∪ (X − A) ⊂ M. Then


X − M ⊂ (X − iκ (A)) ∩ A = cκ (X − A) ∩ A = cκ (X − A) − (X − A). Since X − M is
κ−closed, by Theorem 2.1, X −M = ∅ and so X = M. Conversely, suppose the condition
holds. Let F be a κ−closed set such that F ⊂ A. Since iκ (A)∪(X −A) ⊂ iκ (A)∪(X −F )
and iκ (A) ∪ (X − F ) is κ−open, by hypothesis, X = iκ (A) ∪ (X − F ) and so F ⊂ iκ (A).
By Theorem 2.8, A is gκ −open.

Theorem 2.11. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} and A and B be


subsets of X. If iκ (A) ⊂ B ⊂ A and A is gκ −open, then B is gκ −open.

Proof. The proof follows from Theorem 2.7.

Theorem 2.12. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β}. Then a subset A
of X is gκ −closed if and only if cκ (A) − A is gκ −open.

Proof. Suppose cκ (A) − A is gκ −open. Let A ⊂ M and M be κ−open. Since cκ (A) ∩


(X − M ) ⊂ cκ (A) ∩ (X − A) = cκ (A) − A, cκ (A) − A is gκ −open and cκ (A) ∩ (X − M ) is
κ−closed, by Theorem 2.8, cκ (A) ∩ (X − M ) ⊂ iκ (cκ (A) − A) ⊂ iκ (cκ (A)) ∩ iκ (X − A) =
iκ (cκ (A)) ∩ (X − cκ (A)) = ∅. Therefore, cκ (A) ⊂ M which implies that A is gκ −closed.
S. Maragathavalli, M. Sheik John and D. Sivaraj 61

Conversely, suppose A is gκ −closed. If F ⊂ cκ (A) − A and F is κ−closed, then by


Theorem 2.1, F = ∅ and so F ⊂ iκ (cκ (A) − A). Again, by Theorem 2.8, cκ (A) − A is
gκ −open.

Remark 2.13. Since every topology and quasi-topology are generalized topologies, all
the above theorems are valid in any topological as well as quasi-topological spaces.

3 Igκ −closed sets


Let X be any nonempty set and µ be a GT of subsets of X. A nonempty family I of
subsets of X is said to be a hereditary class [5] if A ∈ I and B ⊂ A, then B ∈ I. Given
a space (X, µ) with a hereditary class I, for each subset A of X, A⋆ (I, µ) = {x ∈ X |
A∩V ̸∈ I for every V ∈ µ such that x ∈ V } [5]. If cµ⋆ (A) = A∪A⋆ (I, µ) for every subset
A of X, then µ⋆ = {A ⊂ X | X − A = cµ⋆ (X − A)} is a GT and cµ⋆ (A) is the intersection
of all µ⋆ −closed supersets of A [5]. iµ⋆ (A) is the largest µ⋆ −open set contained in A. In
this section, we define Igκ −closed sets in a space (X, κ) where κ ∈ {µ, α, σ, π, b, β}
with a hereditary class I and discuss its properties. Throughout the paper, we denote
A⋆ (I, κ) by simply A⋆ . The following Lemma 3.1 will be useful in the sequel.
Lemma 3.1. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary
class I and A and B be subsets of X. Then the following hold.
(a) A⋆ ⊂ B ⋆ whenever A ⊂ B [5, Proposition 2.1].
(b) A⋆ is κ−closed [5, Proposition 2.4].
(c) A is κ⋆ −closed if and only if A⋆ ⊂ A [5, Proposition 3.5].
(d) A⋆ ∪ B ⋆ ⊂ (A ∪ B)⋆ (The proof follows from (a) ).
(e) κ ⊂ κ⋆ [5, Theorem 3.6] and so cκ⋆ (A) ⊂ cκ (A) for every subset A of X and every
κ−closed set is a κ⋆ −closed set.
(f) A ⊂ A⋆ if and only if cκ⋆ (A) = cκ (A) [5, Proposition 3.7].
A subset A of X in a space (X, κ) with a hereditary class I is said to be an Igκ −closed
set if A⋆ ⊂ U whenever U ∈ κ and A ⊂ U. Note that if I = {∅}, then A⋆ (I, κ) = cκ (A) =
cκ⋆ (A) and κ = κ⋆ . So Igκ −closed sets coincide with gκ −closed sets. A is said to be an
Igκ −open set if X − A is an Igκ −closed set. The following Theorems 3.2 and 3.4 give
characterizations of Igκ −closed sets and Theorem 3.3 gives a property of Igκ −closed
sets.
Theorem 3.2. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary
class I. A subset A of X is Igκ −closed if and only if F ⊂ A⋆ − A and F is κ−closed
imply that F = ∅.
Proof. Suppose A is Igκ −closed. Let F be a κ−closed subset of A⋆ −A. Since A ⊂ X −F
and A is Igκ −closed, A⋆ ⊂ X − F and so F ⊂ X − A⋆ . Thus, F ⊂ A⋆ ∩ (X − A⋆ ) = ∅.
Hence F = ∅. Conversely, suppose A ⊂ U and U ∈ κ. Suppose A⋆ ̸⊂ U. Since A⋆ is
κ−closed by Lemma 3.1(b), A⋆ ∩ (X − U ) is a nonempty κ−closed subset of A⋆ − A
which is a contradiction. Therefore, A⋆ ⊂ U and so A is Igκ −closed.
62 On g−closed sets in generalized topological spaces

Theorem 3.3. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary


class I and A be a subset of X. If A is an Igκ −closed subset of X such that A⋆ − A is a
κ−closed set, then A is κ⋆ −closed.

Proof. By Theorem 3.2, A⋆ − A = ∅ and so A⋆ ⊂ A. By Lemma 3.1(c), A is κ⋆ −closed.

Theorem 3.4. If (X, κ) is a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary class


I and A is a subset of X, then the following are equivalent.
(a) A is Igκ −closed.
(b) cκ⋆ (A) ⊂ U whenever A ⊂ U and U is κ−open in X.
(c) For all x ∈ cκ⋆ (A), cκ ({x}) ∩ A ̸= ∅.
(d) cκ⋆ (A) − A contains no nonempty κ−closed set.
(e) A⋆ − A contains no nonempty κ−closed set.

Proof. (a)⇒(b). If A is Igκ −closed, then A⋆ ⊂ U whenever A ⊂ U and U is κ−open in


X and so cκ⋆ (A) = A ∪ A⋆ ⊂ U whenever A ⊂ U and U is κ−open in X. This proves
(b).
(b)⇒(c). Suppose x ∈ cκ⋆ (A). If cκ ({x}) ∩ A = ∅, then A ⊂ X − cκ ({x}). By (b),
cκ⋆ (A) ⊂ X − cκ ({x}), a contradiction, since x ∈ cκ⋆ (A).
(c)⇒(d). Suppose F ⊂ cκ⋆ (A) − A, F is κ−closed and x ∈ F. Since F ⊂ X − A and F
is κ−closed, cκ ({x}) ∩ A = ∅. Since x ∈ cκ⋆ (A) by (c), cκ ({x}) ∩ A ̸= ∅, a contradiction
which proves (d).
(d)⇒(e) is clear.
(e)⇒(a). The proof follows from Theorem 3.2.

From Theorem 3.4(b) and Lemma 3.1(e), it follows that every gκ −closed set is an
Igκ −closed set. The following Example 3.1 shows that an Igµ −closed set need not be
a µ−closed set. Also, since A⋆ is κ−closed, A⋆ is gκ −closed and so A⋆ is Igκ −closed
for every subset A of X. Theorem 3.5 below shows that the concepts gκ −closedness and
Igκ −closedness are equivalent for sets A for which A ⊂ A⋆ .

Example 3.1. Consider the space (X, µ) of Example 2.1 with the hereditary class
I = {∅, {a}, {b} }. Clearly, A = {b} is Igµ −closed but not µ−closed.

Theorem 3.5. If (X, κ) is a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary class


I and A is an Igκ −closed subset of X such that A ⊂ A⋆ , then A is gκ −closed.

Proof. Suppose A is an Igκ −closed subset of X such that A ⊂ A⋆ . If U is a κ−open set


containing A, then by Theorem 3.4(b), cκ⋆ (A) ⊂ U. By Lemma 3.1(f), cκ (A) ⊂ U and
so A is gκ −closed.

Theorem 3.6. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary


class I and A be a subset of X. Then A is Igκ −closed if and only if A = F − N where
F is κ⋆ −closed and N contains no nonempty κ−closed set.
S. Maragathavalli, M. Sheik John and D. Sivaraj 63

Proof. If A is Igκ −closed, then by Theorem 3.4(e), N = A⋆ − A contains no nonempty


κ−closed set. If F = cκ⋆ (A), then F is κ⋆ −closed such that F −N = (A∪A⋆ )−(A⋆ −A) =
(A ∪ A⋆ ) ∩ ((X − A⋆ ) ∪ A) = A. Conversely, suppose A = F − N where F is κ⋆ −closed
and N contains no nonempty κ−closed set. Let U be a κ−open set such that A ⊂ U.
Then F − N ⊂ U which implies that F ∩ (X − U ) ⊂ N. Now A ⊂ F and F ⋆ ⊂ F
imply that A⋆ ∩ (X − U ) ⊂ F ⋆ ∩ (X − U ) ⊂ F ∩ (X − U ) ⊂ N. Since A⋆ is κ−closed by
Lemma 3.1(b), A⋆ ∩ (X − U ) is κ−closed. Therefore, by hypothesis, A⋆ ∩ (X − U ) = ∅
and so A⋆ ⊂ U which implies that A is Igκ −closed.

The following Theorem 3.7 gives a property of Igκ −closed sets and Corollary 3.8
follows from Theorem 3.7 and the fact that, if A ⊂ B ⊂ A⋆ , then A⋆ = B ⋆ and B ⊂ B ⋆ .
Theorem 3.7. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary
class I. If A and B are subsets of X such that A ⊂ B ⊂ cκ⋆ (A) and A is Igκ −closed,
then B is Igκ −closed.
Proof. Since A is Igκ −closed, cκ⋆ (A) − A contains no nonempty κ−closed set. Since
cκ⋆ (B) − B ⊂ cκ⋆ (A) − A, cκ⋆ (B) − B contains no nonempty κ−closed set and so by
Theorem 3.3(d), B is Igκ −closed.

Corollary 3.8. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary


class I. If A and B are subset of X such that A ⊂ B ⊂ A⋆ and A is Igκ −closed, then A
and B are gκ −closed sets.
The following Theorem 3.9 gives a characterization of Igκ −open sets. Theo-
rem 3.10 gives a property of Igκ −open sets which follows from Theorem 3.9.
Theorem 3.9. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary
class I and A be a subset of X. Then A is Igκ −open if and only if F ⊂ iκ⋆ (A) whenever
F is κ−closed and F ⊂ A.
Proof. Suppose A is Igκ −open. If F is κ−closed and F ⊂ A, then X −A ⊂ X −F and so
cκ⋆ (X −A) ⊂ X −F. Therefore, F ⊂ iκ⋆ (A). Conversely, suppose the condition holds. Let
U be a κ−open set such that X − A ⊂ U. Then X − U ⊂ A and so X − U ⊂ iκ⋆ (A) which
implies that cκ⋆ (X − A) ⊂ U. Therefore, X − A is Igκ −closed and so A is Igκ −open.

Theorem 3.10. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary


class I and A ⊂ X. If A is Igκ −open and iκ⋆ (A) ⊂ B ⊂ A, then B is Igκ −open.
The following Theorem 3.11 gives a characterization of Igκ −closed sets in terms
of Igκ −open sets. Theorem 3.12 below shows that every subset of X is Igκ −closed if
and only if every κ−open set is a κ⋆ −closed set.
Theorem 3.11. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary
class I and A ⊂ X. Then the following are equivalent.
(a) A is Igκ −closed.
(b) A ∪ (X − A⋆ ) is Igκ −closed.
(c) A⋆ − A is Igκ −open.
64 On g−closed sets in generalized topological spaces

Proof. (a)⇒(b). Suppose A is Igκ −closed. If U is any κ−open set such that (A ∪ (X −
A⋆ )) ⊂ U, then X − U ⊂ X − (A ∪ (X − A⋆ )) = A⋆ − A. Since A is Igκ −closed, by
Theorem 3.4(e), it follows that X − U = ∅ and so X = U. X may be a κ−open set or
may not be a κ−open set. In both cases, clearly, A ∪ (X − A⋆ ) is Igκ −closed.
(b)⇒(a). Suppose A ∪ (X − A⋆ ) is Igκ −closed. If F is any κ−closed set such that
F ⊂ A⋆ −A, then A∪(X−A⋆ ) ⊂ X−F and X−F is κ−open. Therefore, (A∪(X−A⋆ ))⋆ ⊂
X − F. By Lemma 3.1(d), A⋆ ∪ (X − A⋆ )⋆ ⊂ X − F and so F ⊂ X − A⋆ . Since F ⊂ A⋆ ,
it follows that F = ∅. Hence A is Igκ −closed.
The equivalence of (b) and (c) follows from the fact that X −(A⋆ −A) = A∪(X −A⋆ ).

Theorem 3.12. Let (X, κ) be a space where κ ∈ {µ, α, σ, π, b, β} with a hereditary


class I. Then every subset of X is Igκ −closed if and only if every κ−open set is κ⋆ −closed.

Proof. Suppose every subset of X is Igκ −closed. If U is κ−open, then U is Igκ −closed
and so U ⋆ ⊂ U. Hence U is κ⋆ −closed. Conversely, suppose that every κ−open set is
κ⋆ −closed. If A ⊂ X and U is an κ−open set such that A ⊂ U, then A⋆ ⊂ U ⋆ ⊂ U and
so A is Igκ −closed.

Acknowledgments
The authors thank the referees for their valuable comments and suggestions.

References

[1] Á. Császár. Generalized topology, generalized continuity, Acta Math. Hungar., 2002, 96:
351 – 357.
[2] Á. Császár. γ−connected sets, Acta Math. Hungar., 2003, 100(4): 273 – 279.
[3] Á. Császár. Generalized open sets in generalized topologies, Acta Math. Hungar., 2005,
106: 53 – 66.
[4] Á. Császár. Further remarks on the formula for γ−interior, Acta Math. Hungar., 2006,
113: 325 – 332.
[5] Á. Császár. Modifications of generalized topologies via hereditary classes, Acta Math. Hun-
gar., 2007, 115: 29 – 36.
[6] Á. Császár. Remarks on quasi-topologies, Acta Math. Hungar., 2007, 119: 197 – 200.
[7] N. Levine. Generalized closed sets in topology, Rend. Circ. Mat. Palermo (2), 1970, 19: 89
– 96.
[8] P. Sivagami. Remarks on γ−interior, Acta Math. Hungar., 2007, 119: 81 – 94.

You might also like