1. Introduction
The notion of a
hoop was introduced by Bosbach [
1,
2] as a foundational algebraic structure for examining many-valued logical systems, where the truth values of propositions are represented within a lattice framework. Hoops serve as a valuable mathematical tool for modeling systems in which logical propositions assume a continuum of values, typically within the interval
, rather than being limited to binary truth values. This makes them especially suitable for reasoning in fuzzy logic contexts, where truth is viewed as partial and gradual rather than absolute.
Algebraically, a hoop is a bounded distributive lattice equipped with an additional binary operation, commonly referred to as implication or residuation, that satisfies specific axioms. This structure is particularly significant in fuzzy set theory, where it supports reasoning and decision-making under uncertainty. The implication operation enables the flexible manipulation of fuzzy relations and logical constructs, offering a more expressive framework than classical Boolean logic.
Hoops can be regarded as a generalization of Heyting algebras, extending the semantics of classical logic to accommodate degrees of truth. Their structural properties make hoop algebras indispensable in formalizing the semantics of fuzzy sets and logics and they are widely used in applications such as spatial reasoning, image processing, and decision support systems.
Since Bosbach’s initial formulation, hoop theory has been the subject of extensive research (see [
3,
4,
5,
6,
7,
8,
9,
10,
11]). These studies have deepened our understanding of hoop structures, their properties, and their relevance to many-valued logic systems, particularly within the framework of fuzzy logic.
In 1994, Zhang [
12] introduced the concept of
bipolar fuzzy sets as a generalization of traditional fuzzy sets [
13]. Unlike classical fuzzy sets where membership degrees range from 0 to 1, bipolar fuzzy sets assign degrees within the interval
. In this setting, a degree of 0 indicates irrelevance to the property in question, values between 0 and 1 suggest partial satisfaction, while values between
and 0 indicate the partial satisfaction of a counter-property. Although bipolar fuzzy sets share some conceptual similarities with intuitionistic fuzzy sets, they differ fundamentally in structure and interpretation. The distinction between positive and negative information is crucial in many real-world scenarios; positive information reflects what is considered possible, while negative information captures what is regarded as impossible.
To manage bipolar information, both fuzzy and possibilistic frameworks have been proposed. These approaches are particularly relevant in areas such as image processing and spatial reasoning, where bipolarity naturally arises. For instance, in determining an object’s position, positive information may represent plausible locations, while negative information may signify locations that are definitively excluded. In spatial relation modeling, concepts like “left” and “right” are not mere negations of one another but represent symmetric opposites. This symmetry introduces a middle ground, positions that are neither clearly to the left nor the right, emphasizing the inherent indeterminacy and demonstrating that the union of positive and negative information does not exhaust the entire space.
This paper introduces and investigates the concepts of bipolar fuzzy sub-hoops and bipolar fuzzy filters, examining their algebraic properties in depth. It explores the interactions between these two structures, providing a detailed analysis of their relationship. Additionally, it establishes formal characterizations for both bipolar fuzzy sub-hoops and bipolar fuzzy filters, offering new insights into their underlying structure and contributing to their theoretical development within the broader context of fuzzy logic and many-valued reasoning.
The novelty of this work lies in introducing bipolar fuzzy sub-hoops and filters, which generalize fuzzy logic to incorporate both positive and negative information within hoop algebras. Our contributions include defining these structures, proving their algebraic properties, and characterizing their interrelationship, notably that every BFF is a BFSH (Theorem 4). A key technical difficulty was ensuring that bipolar fuzzy sets respect the hoop algebra’s operations while handling dual membership degrees, which we addressed through new conditions and theorems (e.g., Theorems 3 and 6) that align with classical algebraic structures.
The list of acronyms is given in
Table 1.
2. Preliminaries
Bipolar fuzzy sets, introduced by Zhang [
12], extend traditional fuzzy sets by incorporating negative membership degrees, enabling the modeling of both positive and negative information.
Definition 1 ([
12]).
Let be a non-empty set. A bipolar fuzzy set (briefly, BFS) in is defined as , where and are two functions. The positive membership degree represents the degree to which element satisfies the property associated with the BFS , while the negative membership degree represents the degree to which satisfies an implicit counter-property related to . If and , this indicates that x fully satisfies the positive property of B. Conversely, if and , this suggests that x does not meet the property of but partially satisfies its counter-property. It is also possible that and , which occurs when the membership functions for the property and its counter-property overlap over some subset of . For simplicity, we denote the BFS as . Hoop algebras, defined by Bosbach [
1,
2], provide a lattice-based framework for many-valued logic, with operations ⊙ and → supporting fuzzy reasoning. They generalize Heyting algebras and are central to our bipolar fuzzy extensions.
Definition 2 ([
4]).
A hoop (or hoop algebra) is an algebraic structure denoted as , where forms a commutative monoid, and the following properties hold:(H1) ,
(H2) ,
(H3) for all .
Next, we define a relation
on a hoop
as follows:
It is straightforward to verify that
forms a partially ordered set.
Sub-hoops preserve the algebraic structure of hoops, forming the basis for our bipolar fuzzy sub-hoops, which extend this concept to fuzzy settings.
Definition 3 ([
8]).
A non-empty subset of a hoop algebra is called a sub-hoop of if it satisfies the following condition:Additionally, every sub-hoop must include the element 1. Proposition 1 ([
8]).
Let be a hoop algebra. For any , the following conditions hold:(a1) is a meet-semilattice with ;
(a2) if and only if ;
(a3) , and for any ;
(a4) ;
(a5) and ;
(a6) ;
(a7) ;
(a8) implies , and ;
(a9) .
Filters in hoop algebras capture upward-closed subsets, essential for defining bipolar fuzzy filters that refine fuzzy information.
Definition 4 ([
8]).
A non-empty subset of a hoop algebra is referred to as a filter of if the following conditions are satisfied: Implicative filters strengthen filter conditions, motivating our definition of implicative bipolar fuzzy filters (IBFFs) for advanced fuzzy logic applications.
Definition 5 ([
9]).
A non-empty subset of a hoop algebra is termed an implicative filter of if it satisfies the following conditions: It is important to note that the conditions in (
3) and (4) indicate that the subset
is closed under the operation ⊙ and upward-closed, respectively. Furthermore, a subset
of a hoop algebra
is a filter if and only if it satisfies condition (
5) along with the following additional condition:
3. Bipolar Fuzzy Hoop Algebras
Definition 6. A BFS within a hoop is termed a bipolar fuzzy sub-hoop (briefly, BFSH) of if the following conditions hold:Here, • represents either ⊙ or →. Example 1. Let be a set equipped with two binary operations, · and →, whose values are provided in the following Table 2 and Table 3. Then, is a hoop algebra. Let be a BFS in , given by Table 4. It is straightforward to verify that the BFS within the hoop qualifies as a BFSH of .
Proposition 2. For any BFSH of , the following holds: Proof. Since
for all
, it is straightforward by (
8). □
Proposition 3. Let be a homomorphism of a hoop algebra into a hoop algebra and be a BFSH of . Then, the inverse image of Ω is a BFSH of .
Proof. For any
and
. Then,
and
□
Definition 7. A BFS within a hoop algebra is said to possess the sup-inf property if for any subset , there exists an element such that and .
Proposition 4. Let be a homomorphism from a hoop algebra to a hoop algebra and let be a BFSH of that satisfies the sup-inf property. Then, the image under the homomorphism is a BFSH of .
Proof. For any
and
; let
and
such that
and
Then, by the definition of
, we have
and
Hence,
is a BFSH of
. □
Definition 8. A BFS within a hoop algebra is considered to be a bipolar fuzzy filter (briefly, BFF) of if for every , the following conditions hold: Example 2. Let be a hoop algebra, given in Example 1. Let be a BFS in , given by Table 5. It can be readily verified that the BFS within the hoop algebra qualifies as a BFF of .
The following theorem provides an alternative characterization of bipolar fuzzy filters (BFFs) by replacing the monotonicity condition with a condition involving the implication operation, simplifying their verification in hoop algebras.
Theorem 1. A BFS is a BFF of if and only if it satisfies (9) and Proof. Let
be a BFF of
. Since
for all
, it follows from Equation (11) that Equation (
9) holds. For any
, we obtain the inequality
By utilizing Equations (
10) and (11), we derive the following bounds:
and
which establishes Equation (
12).
Conversely, suppose that the BFS
satisfies both Equations (
9) and (
12). Let
. Since
it follows from the assumptions (
9) and (
12) that
and
Let
be such that
. Then,
, and so
Therefore,
is a BFF of
. □
Theorem 2. In the context of a hoop algebra , every BFF is a BFSH.
Proof. To prove that every BFF is a BFSH in a hoop algebra
, we show that a BFS
satisfying Definition 13 (BFF) also satisfies Definition 6 (BFSH). By Definition 8, a BFF
satisfies
for all
and
, and the monotonicity condition:
Definition 6 requires only the first condition for
. Since this condition is identical in both definitions, it is immediately satisfied by any BFF. The additional monotonicity condition in Definition 8 is not required for a BFSH, so every BFF automatically satisfies the BFSH conditions. Hence, every BFF is a BFSH. □
The converse of Theorem 2 does not necessarily hold, as demonstrated by the following example.
Example 3. Consider the hoop algebra , as defined in Example 1, and the BFSH , presented in the same example. However, does not qualify as a BFF of , as evidenced by the following inequalities:and The following theorem establishes that a bipolar fuzzy set is a BFF if its level sets are filters, bridging fuzzy logic with classical algebraic structures. The proof relies on the definitions of level sets and filter properties in hoop algebras.
Theorem 3. Let be a hoop algebra. The is a BFF of if and only if for all with , the non-empty level sets and are filters of .
Proof. Assume that is a BFF of . Let be such that , and suppose that and are non-empty. It is evident that .
Let be such that and . By the properties of , we have , , and .
From Equation (
12), it follows that
and
Hence,
, which implies that
and
are filters of
.
Conversely, assume that the non-empty level sets and are filters of for all with . For any , let and . Since , we have and .
For any
, let
be such that
,
,
and
. Take
and
. Then, both
and
belong to
. Consequently,
and, thus,
and
Therefore,
is a BFF of
. □
The following theorem characterizes BFFs through conditions on their positive and negative membership functions, enhancing their applicability in fuzzy reasoning systems. The proof uses the algebraic properties of hoop operations to ensure consistency.
Theorem 4. A is a BFF of if and only if it satisfies Equation (9) and the following condition: Proof. Assume that
is a BFF of
and let
. Since
and
, it follows from Equation (11) that
Since
, we also have
and
Thus, by Equations (
12) and (11), we prove (
13).
Conversely, suppose that
satisfies both Equation (
9) and condition (
13). Since
for all
, it follows from (
9) and (
13) that
and
Therefore, by Theorem 1, we conclude that
is a BFF of
. □
The following theorem provides a necessary condition for a bipolar fuzzy set to be a BFF, offering a practical test for filter properties. The proof exploits the interplay between positive and negative memberships under hoop operations.
Theorem 5. A is a BFF of if and only if it satisfies Equation (9) and the following condition: Proof. Assume that
is a BFF of
and let
. Since
, it follows from Equations (11) and (
13) that
This proves Equation (
14).
Conversely, suppose that
satisfies both Equation (
9) and condition (
14). If we set
in (
14), we recover Equation (
12). Thus, by Theorem 1, we conclude that
is a BFF of
. □
The following theorem refines the characterization of BFFs by introducing additional conditions on membership degrees, strengthening their theoretical foundation. The proof combines hoop algebra axioms with bipolar fuzzy set constraints.
Theorem 6. A is a BFF of if and only if it satisfies Equation (9) and the following condition: Proof. Let us assume that
is a BFF of
and consider arbitrary elements
. It is noted that
. Applying Equations (11) and (
13), we obtain the following inequalities:
and
This proves Equation (
15).
Conversely, suppose that
satisfies both Equation (
9) and condition (
15). Setting
in (
15) leads to Equation (
12). Therefore, by Theorem 1, we conclude that
is a BFF of
. □
Theorem 7. A is a BFF of if and only if it satisfies the following condition: Proof. Assume that
is a BFF of
and let
such that
. Since
, it follows that
and
which follows from Equation (
12). Consequently, we obtain the inequalities
and
Conversely, suppose that
satisfies both Equation (
9) and condition (
16). Since for all
, we have
, it follows from (
16) that
Thus, by Theorem 1, we conclude that
is a BFF of
. □
Theorem 8. A is a BFF of if and only if it satisfies (9) and the following condition: Proof. Let us assume that
is a BFF of
. By definition, for all
, the following holds: Since
, for any
, the condition
holds. Therefore, we have the following inequalities:
and, similarly for the negative part,
This confirms that the structure
satisfies the condition in (
17).
Conversely, assume that
satisfies both (
9) and (
17). By substituting
in (
17), we obtain the condition in (
12). Thus, by applying Theorem 1, we conclude that
is indeed a BFF of
. □
Definition 9. A is called an implicative bipolar fuzzy filter (briefly, IBFF) of a hoop algebra if it satisfies the condition given by Equation (9) and Example 4. Let be a set with the binary operations · and → given in the following Table 6 and Table 7. Then, is a hoop algebra. Let be a BFS in given by Table 8. Then, it is easy to check that the BFS in is an IBFF of .
The following theorem introduces implicative bipolar fuzzy filters (IBFFs) as a specialized subset of BFFs, enriching the filter hierarchy.
Theorem 9. Every IBFF is a BFF.
Proof. Let
be an IBFF of
. If we take
in Equation (
18) and use Proposition 1 (a5), then we obtain (
12). Therefore, by Theorem 1,
is a BFF of
. □
The converse of Theorem 9 may not be true, as seen in the following example.
Example 5. Consider the BFF of a hoop algebra in Example 2. However, it is not an IBFF of sinceand Proposition 5. Every IBFF of satisfies the following assertions. Proof. Let
be an IBFF of a hoop algebra
H. If we set
,
and
in Equation (
18) and use Proposition 1 (a5) and (
9), then we obtain (
19). Using (
19), Definition 2 (H1), Proposition 1 (a5), (a7), (a9), and (11), we obtain
and
for all
. It follows from (
9) that we have (20). □
Proposition 6. Let be a bounded hoop algebra. Then, every IBFF of satisfies the following assertions: Proof. Let be an IBFF of a bounded hoop algebra . Then, by Theorem 9, is a BFF of .
If we take
in Equation (20), then
for all
.
Note that
for all
. It follows from Definition 2 (H3) and Equation (11) that
Combining Proposition 1 (a5) and Equations (
9), (
18) and (
24), we obtain
and
This proves Equation (22).
Using (HP3) and Equations (
14) and (22), we have
and
which proves Equation (23). □
Theorem 10. Let be a hoop algebra. The is an IBFF of if and only if its nonempty bipolar fuzzy level sets and are implicative filters of for all with .
Proof. The proof follows similarly to that of Theorem 3. □
We provide conditions under which a BFF is an IBFF.
Theorem 11. If a BFF of satisfies condition (20), then it is an IBFF of .
Proof. Let
. Then, we have
and
This follows from (
9), (
12) and (20). Therefore,
is an IBFF of
. □
Theorem 12. If a BFF of satisfies condition (19), then it is an IBFF of . Proof. Let
be a BFF of
that satisfies (
19). As shown in Proposition 5, (
19) implies (20). Thus, by Theorem 11,
is an IBFF of
. □
Theorem 13. If a BFF of satisfies condition (21), then it is an IBFF of . Proof. Let
be a BFF of
that satisfies (
21). Note that for all
,
By using (
9), (11) and (
21), we obtain the following:
and
Thus, (20) holds and, therefore,
is an IBFF of
by Theorem 11. □
Theorem 14. If a BFF of satisfies condition (22), then it is an IBFF of .
Proof. Let
be a BFF of
that satisfies condition (22). For any
, we have the following:
and, similarly,
These inequalities follow from Definition 2 (H1), (H3), (
9), and (22). Therefore, (
21) holds and, thus, by Theorem 13,
is an IBFF of
. □
Theorem 15. If a BFF of satisfies condition (23), then it is an IBFF of .
Proof. Let
be a BFF of
that satisfies condition (23). Condition (23) implies that
and, similarly,
Thus, (22) is valid and, therefore, by Theorem 14, is an IBFF of . □
Theorem 16. Let and be BFFs of a hoop algebra such thatfor all . If is an IBFF of , then is also an IBFF of . Proof. Assume that is an IBFF of . By Theorem 9, this implies that is a BFF of .
For any
, we have
and
where the equalities follow from (
25), (26) and (
21).
Since
is a BFF of
, we know from (
9) that
Thus, for all
, we conclude
This shows that
satisfies the conditions of an IBFF of
. Therefore, by Theorem 13,
is an IBFF of
. □
Proposition 7. If for all is a family of bipolar fuzzy filters of a hoop algebra , thenis a BFF of . Proof. Let be a family of bipolar fuzzy filters of a hoop .
Let
, and we have
Let
. Then, we have
and
Hence,
is a BFF of a hoop
. □
Lemma 1. Let be a BFF of a hoop algebra if and only if and are fuzzy filters of .
Proof. Assume that
is a BFF of
. Then, by definition,
is a fuzzy filter of
. For every
, we have
Let
. Then, we have
Thus,
is a fuzzy filter of
.
Conversely, assume that
and
are fuzzy filters of
. Then, for every
, we have
and
i.e.,
. Let
. Then, we have
Also,
Thus,
. Therefore,
is a BFF of
. □
Theorem 17. Let be a BFF of if and only if and are fuzzy filters of .
Proof. If a BFS is a BFF of , then by Lemma 1, and are fuzzy filters of . Hence, and are bipolar fuzzy filters of .
Conversely, if and are fuzzy filters of , then and are fuzzy filters of . Therefore, the BFS is a BFF of . □
Definition 10. Let be a BFS in . We define the subset of by Theorem 18. If is a bipolar fuzzy sub-hoop of the hoop , then is a sub-hoop of .
Proof. Clearly,
. Let
. Then,
,
,
, and
. Thus, for
, we have
and
Thus,
and
, which means that
. Therefore,
is a sub-hoop of
. □
A mapping
of hoops is called a homomorphism if
for all
and
. Note that if
is a homomorphism of hoops, then
.
Let
be a homomorphism of hoops. For any BFS
in
, we define a new BFS
in
, where
and
for each
.
For any BFS in , we define a new BFS , .
The following theorem demonstrates that BFFs are preserved under hoop homomorphisms, ensuring their applicability in transformed algebraic systems.
Theorem 19. Let be a homomorphism of hoops. If a BFS is a BFF of , then is a BFF of .
Proof. For all
, we have
Let
. Then,
Hence,
is a BFF of
. □
The following theorem extends the preservation of BFSH properties under homomorphisms, reinforcing their algebraic stability.
Theorem 20. Let be an epimorphism of hoops and let be a BFS in . If is a BFF of , then is a BFF of .
Proof. For any
, there exist
such that
. Then,
Let
. Then, there exist
such that
,
. It follows that
and
Hence,
is a BFF of
. □
Theorem 21. Let be a homomorphism of hoops and be a BFS in . If is a BFF of , then is a BFF of .
Proof. Since
is a homomorphism of
into
, then
and, by the assumption,
for every
. In particular,
for
. Hence,
. Also,
for every
. In particular,
for
. Hence,
. Let
. Then, by assumption,
Hence,
is a BFF of
. □
4. Conclusions
This study advances fuzzy logic by introducing bipolar fuzzy sub-hoops (BFSHs) and bipolar fuzzy filters (BFFs) within hoop algebras, enabling the modeling of both positive and negative information in many-valued logic systems. Below, we summarize the key results and achievements of this study:
Introduced and defined bipolar fuzzy sub-hoops (BFSHs) and bipolar fuzzy filters (BFFs) in hoop algebras, extending fuzzy logic to handle positive and negative memberships.
Established the interplay between BFSHs and BFFs, proving that every BFF is a BFSH (Theorem 2).
Characterized BFFs through level sets (Theorem 3) and alternative conditions involving implication and monotonicity (Theorems 1 and 4).
Defined and characterized implicative bipolar fuzzy filters (IBFFs) as a specialized subset of BFFs (Theorems 10–15).
Proved the preservation of BFSH and BFF properties under hoop homomorphisms, ensuring robustness in transformed systems (Theorems 19–21).
Demonstrated that the intersection of BFFs is a BFF (Proposition 7), supporting their algebraic stability.
Showed that level sets of BFSHs form sub-hoops, connecting fuzzy and classical structures (Theorem 18).
These results find applications in image processing, spatial reasoning, and decision support systems, as detailed in the subsection below. Future research will explore additional algebraic structures, such as bipolar fuzzy rings, and develop computational algorithms for practical applications in image processing, decision-making, and control systems. This work lays a robust foundation for advanced fuzzy logic systems handling complex, contradictory information.
Future work could explore topological indices for fuzzy graphs, such as the fuzzy misbalance prodeg index, which has shown promise in multi-criteria decision-making [new reference]. Applying bipolar fuzzy sub-hoops and filters to such indices could enhance their ability to handle contradictory data in graph-based fuzzy systems.
In image processing, bipolar fuzzy filters enhance edge detection by modeling pixel intensities with positive and negative membership degrees, capturing both desired features (edges) and undesired features (noise). Using the hoop algebra framework, a bipolar fuzzy filter (BFF), as defined in Definition 8, assigns a positive degree
to pixels with high gradient magnitudes, indicating likely edge presence, and a negative degree
to pixels with erratic intensity changes, indicating likely noise. For instance, a pixel at an object boundary may have
and
, while a noisy pixel in a uniform region may have
and
. The BFF processes these memberships using hoop operations ⊙ and →, retaining pixels with high
and low
while suppressing noise, as supported by Theorems 1 and 4 on BFF properties. This dual-membership approach improves segmentation accuracy over traditional fuzzy filters, enabling precise edge detection in applications like medical imaging, where clear tissue boundaries are critical [
12].