1. Introduction
Fuzzy set theory provides a robust framework for addressing uncertainty and imprecision across various domains. Fuzzy sets are employed to represent concepts that are inherently difficult to define precisely, capturing the degree to which elements belong to a set. Over time, this approach has been extended and refined through the integration of various logical systems and algebraic structures. One such system, the strong Sheffer stroke NMV-algebra, represents a generalization of classical Boolean algebra, offering a rich set of operations that are well-suited for modeling fuzzy relations and interactions.
The Sheffer operation, sometimes referred to as the Sheffer stroke or the NAND operator, was first introduced by H. M. Sheffer [
1]. Its significance lies in its ability to operate independently, without reliance on any other logical operations, thereby forming a complete logical system. In other words, all the axioms of a logical framework can be expressed using solely the Sheffer operation, simplifying the analysis and manipulation of these systems. Furthermore, the axioms of Boolean algebra—the algebraic counterpart to classical propositional logic—can be fully articulated using only the Sheffer operation, thus emphasizing its flexibility and utility within logical systems.
The Sheffer stroke operation, due to its unique property of expressing all other logical connectives in Boolean logic, has numerous applications across diverse fields, including computer science, technology, and algebra. In the domain of mathematical logic and algebra, several significant applications have been explored. These include the use of Sheffer stroke operation reducts in basic algebras [
2], an examination of state operators in Sheffer stroke basic algebras [
3], and its role within orthoimplication algebras [
4]. Moreover, the Sheffer stroke operation has been studied within the context of strong non-associative MV-algebras and their filters [
5]. Recent research by Molkhasi et al. has explored representations of strongly algebraically closed algebras, Sheffer stroke algebras, and Visser algebras [
6,
7].
Triangular norms (
-norms) are fundamental operators in fuzzy logic, utilized to model the intersection or conjunction of fuzzy sets. These norms play a critical role in combining degrees of membership within fuzzy set theory, providing a formal mechanism to represent the “and” operation in a fuzzy context.
-norms are defined as binary operations that are associative, commutative, and monotonic, with 1 as the identity element. These properties render
-norms essential in various applications, including decision making, probability theory, and fuzzy logic, where they serve as key tools for managing uncertainty and vagueness in a structured manner. In recent years, numerous studies have investigated the applications of triangular norms [
8,
9,
10,
11,
12].
In this paper, we examine the behavior of fuzzy sets within the context of a strong Sheffer stroke NMV-algebra, with a particular focus on their interaction with -norms. -norms, which generalize the classical concept of intersection in set theory, are critical for understanding the interactions between fuzzy sets. Their application within the strong Sheffer stroke NMV-algebra provides new insights into fuzzy reasoning and logical systems. We investigate the algebraic properties of fuzzy sets under the influence of -norms and explore how these interactions enhance our understanding of uncertainty management within non-classical logical frameworks.
This section reviews the foundational concepts of Sheffer stroke operations [
1], SSNMV-algebras [
13], and t-norms [
14], which are pivotal to our study. The Sheffer stroke, a universal logical operator, facilitates algebraic axiomatization [
2], while SSNMV-algebras generalize non-associative MV-algebras by incorporating strong Sheffer properties [
5]. Triangular norms (t-norms), which model fuzzy intersections, are central to our framework [
8,
14]. Key properties, such as associativity and boundary conditions, are discussed in Lemma 1 [
15]. These tools provide the foundation for our definitions of
-fuzzy subalgebras and filters, effectively bridging algebraic and fuzzy set theories.
The objective of this study is to contribute to the theoretical advancement of fuzzy set theory by examining it through the lens of non-classical algebraic structures. Specifically, we aim to demonstrate how fuzzy sets behave within the strong Sheffer stroke NMV-algebra and how -norms can refine logic and reasoning processes in fuzzy systems. This research opens new avenues for enhanced applications in fields where uncertainty and imprecision are prevalent, such as decision making, artificial intelligence, and fuzzy control systems.
3. Fuzzy Sets with Respect to a t-Norm in Strong Sheffer Stroke NMV-Algebras
In the following sections, let X denote an SSNMV-algebra and represent a -norm, unless otherwise stated. We will introduce the concepts of -fuzzy subalgebras and -fuzzy filters, and explore their properties.
Definition 8. A fuzzy set A in X is called a
-fuzzy subalgebra (briefly, ) of X if -fuzzy filter (briefly, ) of X if
Example 1. Consider the following SSNMV-algebra in Table 1. (i) Let be the product -norm defined by Let A be a fuzzy set in X given by Table 2: Then, it can be easily verified that the fuzzy set A is a -fuzzy subalgebra of X.
(ii) Let denote the Łukasiewicz -norm, which is defined byLet A be a fuzzy set in X given by Table 3: Then, it can be easily verified that the fuzzy set A is -fuzzy subalgebra of X.
(iii) Let represent the product -norm defined by Let A be a fuzzy set in X given by Table 4: Then, it can be easily verified that the fuzzy set A is -fuzzy filter of X.
(iv) Let be the Łukasiewicz -norm defined by Let A be a fuzzy set in X given by Table 5: Then, it can be easily verified that the fuzzy set A is -fuzzy filter of X.
Definition 9. Let A be a fuzzy set in X. Define the subset of X bySince , we have (). Definition 10. A -fuzzy subalgebra (resp. -fuzzy filter) A of X is referred to as an imaginable -fuzzy subalgebra (resp. imaginable -fuzzy filter) of X if A satisfies the imaginable property with respect to .
Example 2. Consider the SSNMV-algebra given in Example 1 of Table 1. (i) Let be the Łukasiewicz -norm defined byLet A be a fuzzy set in X given by Table 6: Then, it can be easily verified that the fuzzy set A is imaginable -fuzzy subalgebra of X.
(ii) Let be the Łukasiewicz -norm defined byLet A be a fuzzy set in X given by Table 7: Then, it can be easily verified that the fuzzy set A is imaginable -fuzzy filter of X.
Proposition 2. If A is an imaginable of X, then .
Proof. Assume that A is an imaginable of X. Let . Then, we have . □
Theorem 2. Let μ be a of X. If there is a sequence in X such that , then .
Proof. Let
. Then, we have
Hence, for any
, we have
Since
, we conclude that
□
Theorem 3. If A is an imaginable of X, then is an SSNMV-subalgebra of X.
Proof. Assume that
A is an imaginable
of
X. Let
. Then, we have
and
. Thus, we obtain the following inequality:
Thus, we conclude that
which implies that
. Therefore,
is an SSNMV-subalgebra of
X. □
Theorem 4. If A is a of X with , then is an SSNMV-filter of X.
Proof. Assume that
A is a
of
X with
. By Definition 9, we know that
. Let
be such that
and
. Then, by the definition of
, we have
and
. Thus, we can write:
Thus, we conclude that
, which implies that
. Therefore,
is an SSNMV-filter of
X. □
Let
S be a nonempty subset of
X, and let
with
. The
-characteristic function
of
X is a function from
X into
, defined as follows:
By this definition, the
-characteristic function
maps elements of
X into
. The fuzzy set
in
X, associated with
, is called the
-characteristic fuzzy set of
S in
X.
Lemma 2. Let S be a nonempty subset of X. Then, the following statements hold:
- 1.
If the constant 1 of X belongs to S, then for all , we have .
- 2.
If there exists an element such that , then the constant 1 of X must be an element of S.
Proof. (1) If , then for all .
(2) Assume that there exists an element such that . Then, , so it follows that . Hence, the constant element . □
Theorem 5. If S is an SSNMV-subalgebra of X, then is a of X.
Proof. Assume that S is an SSNMV-subalgebra of X.
Case 1: Let
. Then,
. Since
S is a subalgebra of
X, we have
, and consequently,
. By Lemma 1 (1), we obtain
Case 2: Let
or
. In this case, either
or
. By Lemma 1 (1), we have
Therefore,
is a
of
X. □
Theorem 6. Let S be a nonempty subset of X such that is a of X and satisfies . Then, S is an SSNMV-subalgebra of X.
Proof. Assume that
S is a nonempty subset of
X such that
is a
of
X with
. Let
. Then, by the definition of the characteristic function, we have
By the properties of the
, specifically using (
1), we obtain the following:
Therefore, we conclude that
which implies that
. Hence,
S is an SSNMV-subalgebra of
X. □
If , then Theorem 6 is not applicable, as demonstrated by the following example.
Example 3. Consider the SSNMV-algebra defined in Example 1 of Table 1. Let , and define a fuzzy set in X as given by the table in Table 8: Let be the Łukasiewicz -norm, defined asThen, is a -fuzzy subalgebra of X. However, we observe thatTherefore, the condition does not hold, and Theorem 6 cannot be applied. Additionally, S is not an SSNMV-subalgebra of X because , but we have Corollary 1. A nonempty subset S of X is an SSNMV-subalgebra of X if and only if is a of X.
Proof. The result follows directly from Theorems 5 and 6. □
Theorem 7. Let P be an SSNMV-subalgebra of X and A be a fuzzy set in X defined byfor all with . Then, A is a -FS of X, where is the -norm defined byIn particular, if and , then A is an imaginable -fuzzy subalgebra of X. Moreover, . Proof. Let . We consider three cases:
Case i: If
, then
Case ii: If
and
(or,
and
), then
Case iii: If
, then
Hence, A is a -fuzzy subalgebra of X.
Assume that
and
. Then,
and
Thus,
, that is,
. Therefore,
A is an imaginable
-fuzzy subalgebra of
X. Also,
Hence,
. □
Theorem 8. If S is an SSNMV-filter of X, then is a of X.
Proof. Assume that S is an SSNMV-filter of X. Since , it follows from Lemma 2 (1) that for all . Next, let .
Case 1: If
and
, then
Since
S is an SSNMV-filter of
X, we have
and thus
. By Lemma 1 (1), we have
Case 2: If
or
, then
By Lemma 1 (1), we have
Therefore,
is a
of
X. □
Theorem 9. If S is a nonempty subset of X such that is a of X with , then S is an SSNMV-filter of X.
Proof. Assume that S is a nonempty subset of X such that is a of X with . Then, by Lemma 2 (2), we have for all , which implies that .
Let
be such that
and
. Then, we have
By Equation (3), we obtain
Therefore,
, which implies that
. Hence,
S is an SSNMV-filter of
X. □
If , then Theorem 9 is not applicable, as demonstrated in the following example:
Example 4. Consider the SSNMV-algebra given in Example 1 of Table 1. Let . Define a fuzzy set in X by the table in Table 9: Let be the Lukasiewicz -norm defined byThen, is a -fuzzy filter of X. However, . Furthermore, S is not an SSNMV-filter of X, since , but while . Corollary 2. A nonempty subset S of X is an SSNMV-filter of X if and only if is a of X.
Proof. The result follows immediately from Theorems 8 and 9. □
4. Level Subsets of a Fuzzy Set with Respect to a -Norm
Fuzzy sets have become a fundamental concept in various fields, and the analysis of their level subsets plays a crucial role in understanding their properties and applications. This section specifically focuses on the level subsets of a fuzzy set with respect to a -norm, a mathematical framework that enables a more nuanced analysis of fuzzy relations and operations.
Definition 11. Let A be a fuzzy set in a nonempty set X. For any , we define the following sets:These sets are referred to as the upper -level subset, lower -level subset, and equal -level subset of A, respectively. For a fuzzy set A in X, it follows that , , and . Lemma 3. Let A be a fuzzy set in X. The following statements hold:
- 1.
If the constant element 1 of X belongs to , then for all .
- 2.
For any , if there exists an element such that , then the constant element 1 of X is in .
Proof. (1) If , then for all .
(2) Suppose that for some . Since , it follows that . Therefore, we have , which implies that . □
Theorem 10. If A is a of X, then for every such that , the set is an SSNMV-subalgebra of X whenever it is nonempty. In particular, is an SSNMV-subalgebra of X if is nonempty.
Proof. Let
A be a
of
X. Consider any
such that
, and assume that
. Let
. Then, by definition, we have
and
. Therefore,
Hence,
, implying that
is an SSNMV-subalgebra of
X. □
Theorem 11. If a fuzzy set A in X is such that for every , the set is an SSNMV-subalgebra of X whenever it is nonempty, then A is a of X.
Proof. Suppose that for each , the set is an SSNMV-subalgebra of X whenever it is nonempty. We aim to show that A is a of X.
Assume, for the sake of contradiction, that there exist
such that
Define
Then,
, and we have the inequalities
This implies that
.
By Lemma 3 (1), since , we know that if , then is nonempty.
Hence, the assumption that leads to a contradiction, as it implies that cannot be an SSNMV-subalgebra of X, contradicting our assumption that all nonempty are SSNMV-subalgebras.
Therefore, we conclude that for all , which implies that A is a of X. □
The converse of Theorem 11 need not be true, as demonstrated in the following example.
Example 5. Consider the SSNMV-algebra given in Example 1 of Table 1. Let be the Łukasiewicz -norm, defined by Let A be a fuzzy set in X, as specified in Table 10: The fuzzy set A is a -fuzzy subalgebra of X. Let . Then, the upper level subset isHowever, is not an SSNMV-subalgebra of X. This is because, while , we have the following:Therefore, the set is not closed under the operations required for it to be an SSNMV-subalgebra of X, thereby illustrating that the converse of Theorem 11 does not hold in this case. Theorem 12. Let A be a of X. For every such that , the set is an SSNMV-filter of X whenever it is nonempty. In particular, if is nonempty, then is an SSNMV-filter of X.
Proof. Assume that A is a of X. Let such that , and suppose that . Since A is a of X, we know that for all . By Lemma 3 (2), we conclude that .
Next, let
be such that
and
. From the assumption, we know that
By the property of the
-norm, we have
Therefore,
, which shows that
is an SSNMV-filter of
X. □
Theorem 13. Let A be a fuzzy set in X. If for all , the set is an SSNMV-filter of X whenever it is nonempty, then A is a of X.
Proof. Assume that for every , the set is an SSNMV-filter of X whenever it is nonempty. For any , let . Then, . By assumption, is an SSNMV-filter of X, so . This implies that .
Now, assume that there exist
such that
Define
Thus,
and we have
This implies that
. By Lemma 3 (1), we know that
. Therefore,
cannot be an SSNMV-filter of
X, which leads to a contradiction, since
was assumed to be an SSNMV-filter of
X. Hence, we conclude that
for all
. Therefore,
A is a
of
X. □
The converse of Theorem 13 need not be true, as shown in the following example.
Example 6. Consider the SSNMV-algebra given in Example 1 of Table 1. Let be the Łukasiewicz -norm defined byLet A be a fuzzy set in X given by the Table 11: Then, the fuzzy set A is a -fuzzy filter of X. If , then . However, is not an SSNMV-filter of X sinceand Definition 12. Let be a function, where X and Y are nonempty sets. If B is a fuzzy set in Y, the inverse image of B under is a fuzzy set in X, denoted by . This fuzzy set is defined as follows:where denotes the membership function of the fuzzy set B in Y. Definition 13. Let and be SSNMV-algebras. A mapping is called a homomorphism ifA homomorphism is called an epimorphism if is surjective. In the following, let represent a homomorphism from an SSNMV-algebra to an SSNMV-algebra , unless stated otherwise. It is important to note that for every , the condition holds.
Theorem 14. Let be a surjective homomorphism between SSNMV-algebras. If B is a - in Y, then the inverse image of B under , denoted by , is a in X.
Proof. Let
be a surjective homomorphism, and assume that
B is a
in
Y. For any
, we observe the following:
Thus,
is a
in
X. □
Theorem 15. Let be a surjective homomorphism between SSNMV-algebras. If B is a in Y such that the mapping is order-preserving, then is a in X.
Proof. Let
be a surjective homomorphism, and suppose that
B is a
in
Y with
being an order-preserving mapping. For every
, we have
Now, for any
, we observe
Thus,
is a
in
X. □
Definition 14 ([
17])
. Let X and Y be nonempty sets, and let be a function. A fuzzy set A in X is said to be invariant under if the following condition holds for all : Definition 15. Let be a function from a nonempty set X to a nonempty set Y. For a fuzzy set A in X, the image of A under , denoted , is a fuzzy set in Y defined as follows: Definition 16 ([
18])
. A fuzzy set A in X has the sup property if for any nonempty subset S of X, there exists such that Lemma 4 ([
19])
. Assume that is surjective. Let A be a -invariant fuzzy set in X with the sup property. For any , there exist and such that and . Theorem 16. Let be a surjective function. If A is a in X that is invariant under and satisfies the sup property, then the image of A under , denoted by , in Y is also a .
Proof. Let
A be a
-invariant
in
X that satisfies the sup property. Consider
. Since
is surjective, the inverse images
and
are nonempty subsets of
X. By Lemma 4, there exist elements
and
such that
and
. Thus,
Hence,
is a
in
Y. □
Theorem 17. Let be a surjective function. If A is a in X that is -invariant and possesses the sup property, then the image of A under , denoted , is a in Y.
Proof. Let
A be a
-invariant
in
X that satisfies the sup property. Since
A is a
, it follows that
for every
. Given that
, we know that
, implying that
is nonempty. Therefore, we have
Thus, we obtain
Next, for
, by Lemma 4, there exist
and
such that
and
. Thus, we have
Hence,
is a
in
Y. □
Definition 17 ([
20])
. Let be a nonempty collection of fuzzy sets in a non-empty set X. The intersection of the family in X is defined by the membership function as follows: Similarly, the union of the family in X is defined by its membership function as follows: Theorem 18. Let be a nonempty collection of -fuzzy subalgebras of X. Then, the intersection is also a of X.
Proof. Let
. For each
, we know that
Since for each
, the infimum of the membership functions satisfies
applying Lemma 1 (3) yields
for all
. Therefore, we obtain the following inequality:
Thus, we conclude that
is a
of
X. □
Theorem 19. Let and be -fuzzy subalgebras of the sets X and Y, respectively. Then, the Cartesian productwhere the membership function is defined asfor all is a of . Proof. Let
. Then,
Hence,
is a
of
. □
Theorem 20. Let and be -fuzzy filters of the sets X and Y, respectively. Then, the productis a of , where the membership function is given byfor every pair . Proof. Let
. Then,
Let . Then,
Hence,
is a
of
. □
Theorem 21. Consider a descending chain of SSNMV-filters on X that eventually terminates after a finite number of steps. Let A be a on X. If the sequence of values in the image of is strictly increasing, then A must be finite-valued.
Proof. Assume, for the sake of contradiction, that A is infinite-valued. Let denote a strictly increasing sequence of values from the image of , such that . Consider the SSNMV-filter on X for each .
If
for
, it follows that
, which implies that
. Thus, we obtain the inclusion
Since
is in the image of
, there exists some
such that
. Therefore,
, but
. This implies the strict inclusion:
Consequently, we obtain a strictly descending chain of SSNMV-filters on
X:
which does not terminate. This is a contradiction to the assumption that the chain eventually terminates. Therefore,
A must be finite-valued. □
Theorem 22. If every on X has a finite image, then any descending sequence of SSNMV-filters of X must eventually terminate after a finite number of steps.
Proof. Assume that there exists an infinite strictly descending chain of SSNMV-filters on
X:
which does not terminate after a finite number of steps. Now, define a fuzzy set
on
X as follows:
where
, and
. This construction defines
as a fuzzy set on
X where the membership values decrease as the SSNMV-filters decrease in the sequence.
Now, assume that . Suppose that for some and for some , with without loss of generality.
Clearly, since
, we have
because
is an SSNMV-filter. Hence, we have:
If
, then
, and thus,
If
but
, there exists an integer
such that
. This implies that
, and therefore,
Lastly, if
but
, there exists an integer
such that
, leading to:
Therefore, we conclude that A is a with an infinite number of distinct values, which contradicts the assumption that every on X has a finite image. Hence, the assumption of an infinite strictly descending chain of SSNMV-filters must be incorrect, and the chain must eventually terminate after a finite number of steps. □
Theorem 23. Let be a -norm defined on I, and let μ be a fuzzy set in an SSNMV-algebra X with the image of μ given by , where for . Assume that there exists a chain of SSNMV-subalgebras of X:such that for each , , where and . Then, μ is a on X. Proof. Let
. If both
and
are elements of the same
, then we have
and
. This leads to
Now, suppose that
and
for some
. Without loss of generality, assume that
. In this case,
and
, where
. Since
, we have
Thus, we conclude that is a on X. □
5. Conclusions
In this paper, we explore the concept of fuzzy sets in the context of triangular norms, specifically applied to strong Sheffer stroke NMV-algebras. We introduce the notions of -fuzzy subalgebras and -fuzzy filters, and investigate their properties, providing several illustrative examples. Additionally, we examine the relationships between -fuzzy subalgebras (filters) and their classical counterparts, as well as their level subsets and homomorphisms.
The results presented contribute to a deeper understanding of fuzzy structures within NMV-algebras and their potential applications. The interplay between fuzzy subalgebras, filters, and classical counterparts opens new avenues for further research.
Theoretical Implications: This paper generalizes fuzzy set theory to non-classical algebraic structures, namely SSNMV-algebras, thereby enriching the mathematical foundation of fuzzy logic. The introduction of -fuzzy subalgebras and -fuzzy filters provides a unified framework for studying algebraic and logical properties under uncertainty.
Practical Implications: This work enhances tools for artificial intelligence (AI) and decision making by modeling imprecise logical operations (e.g., via Sheffer stroke and -norms). It offers potential applications in areas such as control systems, database systems, and multi-valued logic design, where uncertainty management is crucial.
Future work could focus on extending the concepts introduced in this paper to more general classes of algebras, particularly those arising in other areas of mathematics and logic. Specifically, investigating the implications of -fuzzy subalgebras and -fuzzy filters in the context of multi-valued logic or lattice theory represents a promising direction. Furthermore, exploring the computational aspects of these structures—such as algorithms for constructing or analyzing -fuzzy subalgebras and filters—could provide valuable practical insights. Additionally, the study of homomorphisms between fuzzy algebras and their applications in optimization or decision-making processes presents an exciting area for future research.