Covering-based fuzzy rough sets

Abstract

Many researchers have combined rough set theory and fuzzy set theory in order to easily approach problems of imprecision and uncertainty. Covering-based rough sets are one of the important generalizations of classical rough sets. Naturally, covering-based fuzzy rough sets can be studied as a combination of covering-based rough set theory and fuzzy set theory. It is clear that Pawlak’s rough set model and fuzzy rough set model are special cases of the covering-based fuzzy rough set model. This paper investigates the properties of covering-based fuzzy rough sets. In addition, operations of intersection, union and complement on covering-based fuzzy rough sets are investigated. Finally, the corresponding algebraic properties are discussed in detail.

Keywords

Fuzzy rough sets monotone covering approximation operator algebraic property

1 Introduction

Rough set theory, first proposed by Pawlak [35], is an excellent tool with which to handle vagueness and uncertainty in data analysis. The theory has been applied to the fields of medical diagnosis, conflictanalysis, pattern recognition and data mining [7 , 19].

Pawlak rough set theory is built on equivalence relations. However, an equivalence relation is restrictive for many real-world applications [8 , 22]. To overcome this limitation, there are two primary methods to generalize Pawlak rough set theory. Rough set theory has been generalized from the perspective of extending the equivalence relation to other binary relations, such as dominance relations, tolerance relations and similarity relations [15, 32]. In addition, one of the most important generalizations is to replace a partition obtained by the equivalence relation with a covering [3 , 34]. Zakowski, in 1983, first employed the covering of a universe to establish a covering-based generalized rough set. Since then, the study of covering-based rough set theory has attracted many researchers. Many kinds of lower and upper approximation operators have been proposed [18 , 31]. Yao proposed approximation operators based on coverings produced by the predecessor and/or successor neighborhoods of serial or inverse serial binary relations [33]. Zhu, et al. studied six types of approximation operators and investigated the properties and relationships among them [26 –28]. Qian, et al. simultaneously investigated five pairs of dual covering-based approximation operators by employing the notion of the neighborhood [13]. In addition, Yun, et al., also discussed covering rough sets and solved an open problem identified by Zhu and Wang [27]. To construct the lower and upper approximations of an arbitrary, Chen, et al. proposed a new covering-based on generalized rough set [3].

Alternatively, rough set theory was generalized by combining with other theories that deal with uncertain knowledge. The fuzzy rough set model which combines fuzzy set theory with rough set theory is one of the most important adaptations. It is well known that fuzzy set theory and rough set theory are complementary in terms of handling different kinds of uncertainty. Rough set theory deals with uncertainty resulting from ambiguity of information [1], while fuzzy set theory is adept at dealing with probabilistic uncertainty, connected to the imprecision of states, perceptions and preferences. The two theories can be encountered in many specific problems. Therefore, rough set theory has been generalized by combining it with fuzzy set theory. Many researchers have discussed the fuzzy rough set model from various perspectives [1 , 21]. Dubois and Prade proposed the concepts of the rough fuzzy set and the fuzzy rough set [2]. Morsi, et al. discussed some axioms of fuzzy rough sets [16]. Wu, et al. studied the (I, T)-fuzzy rough approximation operators [20]. Xu, et al. proposed the multi-granulation fuzzy rough set model and studied the properties of multi-granulation fuzzy rough sets [23]. However, it is still an open problem regarding the research of covering-based fuzzy rough sets.

In this paper, the primary objective is to investigate covering-based rough set theory when combined with fuzzy set theory. The paper is organized as follows. In Section 2, some basic concepts of Pawlak’s rough set theory and fuzzy rough set theory are described. Furthermore, the concept of the monotone covering is proposed. In Section 3, the properties of covering-based fuzzy approximation operators are investigated. In Section 4, the operations of intersection, union and complement on covering-based fuzzy rough sets are discussed, as are the algebraic properties of covering-based fuzzy rough sets. Finally, Section 5 concludes this study.

2 Preliminaries

In this section, some basic concepts and notions according to Pawlak’s theory rough sets, fuzzy sets, and covering are described. Additional details can be found in various references [23 , 35].

(U, R) is referred to as an approximation space, in which U = {x₁, x₂, ⋯ , x_n} is a non-empty finite set. $ℝ = {R_{1}, R_{2}, \dots, R_{m}}$ is a set of the equivalence relations. Denote [x] _R = {y| (x, y) ∈ R}, U/R = {[x] _R|x ∈ U}; then, [x] _R is called the equivalence class of x and the quotient set U/R is called the equivalence class set of U.

Definition 2.1. Let $(U, ℝ)$ be an approximation spaceand R be an equivalence relation. For any X ⊆ U, $\underline{R} (X) = {x \in U | [x]_{R} \subseteq X}$ , $\bar{R} (X) = {x \in U | [x]_{R} \cap X \neq$ φ}.

These are the Pawlak lower and upper approximations of X with respect to equivalence relation R, respectively.

Let U represent a non-empty finite set. A fuzzy set X is a mapping from U into the unit interval [0, 1]; X : U → [0, 1], where each X (x) is the membership degree of x in X. The set of all the fuzzy sets defined on U is denoted by F (U).

Definition 2.2. Let $(U, ℝ)$ be an approximation space and R be an equivalence relation. For any X ⊆ U, denote $\begin{matrix} \underline{R} (X) (x) = \land {X (y) | y \in [x]_{R}} \\ \bar{R} (X) (x) = \lor {X (y) | y \in [x]_{R}} . \end{matrix}$

$\underline{R} (X)$ and $\bar{R} (X)$ are the lower and upper approximations of the fuzzy set X with respect to equivalence relation R, where ∧ represents “min” and ∨ represents “max”.

Definition 2.3. Let U be the universe and $ℂ$ a family of nonempty subsets of U. If $\cup ℂ = U$ , then $ℂ$ is a covering of U. The ordered pair $(U, ℂ)$ represents a covering approximation space.

Let $(U, ℂ)$ be a covering approximation space. For any $x \in U, {K_{x} \in ℂ | x \in K_{x}}$ is denoted as $st (x, ℂ)$ , i.e., $st (x, ℂ) = {K_{x} \in ℂ | x \in K_{x}}$ .

Definition 2.4. Let $ℂ$ be a covering of U. For anyx ∈ U and $st (x, ℂ) = {K_{x_{1}}, K_{x_{2}}, \dots, K_{x_{n}}}$ . If K_{x
₁}, K_{x
₂}, ⋯ , K_{x
_n} can be reordered to K_{x
_i1}, K_{x
_i2}, ⋯ , K_{x
_in} such that K_{x
_i1} ⊆ K_{x
_i2} ⊆ ⋯ K_{x
_in}, then $ℂ$ is a monotone covering of U, and $(U, ℂ)$ is a monotone covering approximation space.

Furthermore, we denote $K_{x}^{min} = K_{x_{i 1}}$ and $ℂ^{min} =$ ${K_{x}^{min} | x \in U}$ . In addition, denote, $ℂ^{MIN} = {K_{x_{i}}^{min},$ i = 1, 2, ⋯ , m}, where $ℂ^{MIN}$ must satisfy the following three conditions.

(1) For any $K_{x_{i}}^{min} \in ℂ^{MIN}$ , $K_{x_{i}}^{min} \in ℂ^{min}, i = 1,$ 2, ⋯ , m ;

(2) For any $K_{x_{i}}^{min}, K_{x_{j}}^{min} \in ℂ^{MIN}, K_{x_{i}}^{min} \cap K_{x_{j}}^{min} = φ,$ i≠ j, i, j = 1, 2, ⋯ , m ;

(3) For any $K_{x}^{min} \in ℂ^{min}$ , there exists $K_{x_{i}}^{min} \in ℂ^{MIN}$ such that $K_{x_{i}}^{min} \subseteq K_{x}^{min} .$

Remark 2.1. For any $K \in ℂ^{MIN}$ and any x ∈ K, by the construction of $ℂ^{MIN}$ , $K = K_{x}^{min}$ can be easily obtained.

Example 2.1. Let U = {x₁, x₂, ⋯ , x₈} , K₁ = {x₁, x₂, x₃, x₄, x₅} , K₂ = {x₃, x₄, x₅} , K₃ = {x₅} , K₄ = {x₆, x₇, x₈} , and K₅ = {x₈} , $ℂ = {K_{1}, K_{2}, K_{3}, K_{4}, K_{5}} .$ Then $ℂ$ is a monotone covering of U. Moreover, $ℂ^{min} = {{x_{1}, x_{2}, x_{3}, x_{4}, x_{5}}, {x_{3}, x_{4}, x_{5}}, {x_{5}}, {x_{6}, x_{7}, x_{8}}, {x_{8}}}$ , $ℂ^{MIN} = {{x_{5}}, {x_{8}}} .$

Definition 2.5. Let $(U, ℂ)$ be a covering approximation space. For any fuzzy set X ∈ F (U), denote $\begin{matrix} \underline{C} (X) (x) = \lor_{K_{x} \in ℂ} {\land {X (y) | y \in K_{x}}} \\ \bar{C} (X) (x) = \land_{K_{x} \in ℂ} {\lor {X (y) | y \in K_{x}}} . \end{matrix}$

$\underline{C} (X)$ and $\bar{C} (X)$ are the lower and upper covering fuzzy approximations of X, respectively. The pair $(\underline{C} (X), \bar{C} (X))$ is the covering fuzzy rough set of X and $\tilde{C} = {(\underline{C} (X), \bar{C} (X)) | X \in F (U)}$ represent all of the covering fuzzy rough sets.

Remark 2.2. In a special case, when $ℂ$ is a partition of the universe, then $\underline{C} (X)$ and $\bar{C} (X)$ of Definition 2.5 will degenerate into $\underline{R} (X)$ and $\bar{R} (X)$ of Definition 2.2.

Example 2.2. Let U = {x₁, x₂, x₃, x₄} , K₁ = {x₁, x₂}, K₂ = {x₂} , K₃ = {x₁, x₃} , K₄ = {x₃, x₄}, and $ℂ = {K_{1}, K_{2}, K_{3}, K_{4}} .$ $ℂ$ is a covering of U. For fuzzy set X = (0.3, 0.4, 0.2, 0.5) , $\underline{C} (X) = (0.3, 0.4, 0.2, 0.5),$ $\bar{C} (X) = (0.3, 0.4, 0.3, 0.5) .$

3 Covering-based fuzzy approximation operators

In the section, the properties of the lower and upper covering fuzzy approximation operators in a covering approximation space are considered.

Proposition 3.1.Let $(U, ℂ)$ be a covering approximation space andX, Y ∈ F (U). Then, the following properties hold: $\begin{matrix} (1) \underline{C} (U) = \bar{C} (U) = U, \underline{C} (φ) = \bar{C} (φ) = φ; \\ (2) \underline{C} (X) \subseteq X \subseteq \bar{C} (X); \\ (3) X \subseteq Y \Rightarrow \underline{C} (X) \subseteq \underline{C} (Y), \bar{C} (X) \subseteq \bar{C} (Y); \\ (4) \underline{C} (\sim X) = \sim \bar{C} (X), \bar{C} (\sim X) = \sim \underline{C} (X) . \end{matrix}$

Proof: The properties can be easily proved by Definition 2.5.□

Example 3.1. (Continued from Example 2.2) Foranother fuzzy set Y = (0.3, 0.2, 0.4, 0.1) , $\underline{C} (X) \cap$ $\underline{C} (Y) = (0.3, 0.2, 0.2, 0.1) .$ $\underline{C} (X \cap Y) = (0.2, 0.2, 0.2,$ 0.1) . $\underline{C} (X) \cap \underline{C} (Y) \neq \underline{C} (X \cap Y)$ . Similarly, $\bar{C} (X) \cup \bar{C} (Y) \neq \bar{C} (X \cup Y) .$

If $ℂ$ is a monotone covering of U, then the following properties are observed.

Proposition 3.2.Let $(U, ℂ)$ be a monotone covering approximation space andX, Y ∈ F (U). Then the following properties hold: $\begin{matrix} (1) \underline{C} (X \cap Y) = \underline{C} (X) \cap \underline{C} (Y); \\ (2) \bar{C} (X \cup Y) = \bar{C} (X) \cup \bar{C} (Y) . \end{matrix}$

Proof: (1) (⇒ :) can be proved easily by Proposition 3.1. $\begin{matrix} (\Leftarrow :) For \forall x \in U, we have \underline{C} (X \cap Y) (x) \\ = \lor_{K_{x} \in ℂ} {\land {(X \cap Y) (y) | y \in K_{x}}} \\ = \land {(X \cap Y) (y) | y \in K_{x}^{min}} \\ \geq {\land {X (y) | y \in K_{x}^{min}}} \land {\land {Y (y) | y \in K_{x}^{min}}} \\ = \underline{C} (X) (x) \land \underline{C} (Y) (x) \\ = (\underline{C} (X) \cap \underline{C} (Y)) (x) \end{matrix}$

(2) The item can be proved similarly to (1).□

Proposition 3.3.Let $(U, ℂ)$ be a monotone covering approximation space andX ∈ F (U). Then the following properties hold: $\begin{matrix} (1) \underline{C} (\underline{C} (X)) = \underline{C} (X); \\ (2) \bar{C} (\bar{C} (X)) = \bar{C} (X) . \end{matrix}$

Proof: (1) (⇒ :) can be proved easily by Proposition 3.1. $\begin{matrix} (\Leftarrow :) For any x \in U, we have \underline{C} (\underline{C} (X)) (x) \\ = \lor_{K_{x} \in ℂ} {\land {\underline{C} (X) (y) | y \in K_{x}}} \\ = \land {\underline{C} (X) (y) | y \in K_{x}^{min}} \\ = \underline{C} (X) (y_{0}) (where \underline{C} (X) (y_{0}) = min_{y \in K_{x}^{min}} \underline{C} (X) (y)) \\ = \lor_{K_{y_{0}} \in ℂ} {\land {X (z) | z \in K_{y_{0}}}} \\ = \land {X (z) | z \in K_{y_{0}}^{min}} \\ \geq \land {X (z) | z \in K_{x}^{min}} \\ = \lor_{K_{x} \in ℂ} {\land {X (z) | z \in K_{x}}} \\ = \underline{C} (X) (x) \end{matrix}$

(2) The item can be proved similarly to (1). □

Proposition 3.4.Let $(U, ℂ)$ be a monotone covering approximation space andX, Y ∈ F (U). Then the following properties hold: $\begin{matrix} (1) \underline{C} (\underline{C} (X) \cap \underline{C} (Y)) = \underline{C} (X) \cap \underline{C} (Y); \\ (2) \underline{C} (\underline{C} (X) \cup \underline{C} (Y)) = \underline{C} (X) \cup \underline{C} (Y); \\ (3) \bar{C} (\bar{C} (X) \cap \bar{C} (Y)) = \bar{C} (X) \cap \bar{C} (Y); \\ (4) \bar{C} (\bar{C} (X) \cup \bar{C} (Y)) = \bar{C} (X) \cup \bar{C} (Y) . \end{matrix}$

Proof: (1) It is clear according to Proposition 3.1 and Proposition 3.2.

(2) (⇒ :) is clear according to Proposition 3.1.

$\begin{matrix} (\Leftarrow :) For \forall x \in U, \underline{C} (\underline{C} (X) \cup \underline{C} (Y)) (x) \\ = \lor_{K_{x} \in ℂ} {\land {(\underline{C} (X) \cup \underline{C} (Y)) (y) | y \in K_{x}}} \\ = \land {(\underline{C} (X) \cup \underline{C} (Y)) (y) | y \in K_{x}^{min}} \\ \geq {\land {\underline{C} (X) (y) | y \in K_{x}^{min}}} \lor {\land {\underline{C} (Y) (y) | y \in K_{x}^{min}}} \\ = \underline{C} (\underline{C} (X)) (x) \lor \underline{C} (\underline{C} (Y)) (x) \\ = \underline{C} (X) \cup \underline{C} (Y) (x) . \end{matrix}$

(3) This item can be proved similarly to (2).

(4) This item can be proved similarly to (1). □

Proposition 3.5.Let $(U, ℂ)$ be a monotone covering approximation space . For anyx ∈ UandX ∈ F (U), if $K_{x}^{min} = {x}$ , then $\underline{C} (X) (x) = \bar{C} (X) (x) .$

Proof: Clear according to Definition 2.5.□

Proposition 3.6.Let $(U, ℂ)$ be a monotone covering approximation space. For ∀x, y ∈ UandX ∈ F (U), if $K_{x}^{min} = K_{y}^{min}$ , the following properties hold: $\begin{matrix} (1) \underline{C} (X) (x) = \underline{C} (X) (y); \\ (2) \bar{C} (X) (x) = \bar{C} (X) (y) . \end{matrix}$

Proof: Clear according to Definition 2.5.□

4 Covering-based fuzzy rough sets

4.1 Operations on covering-based fuzzy rough sets

In this section, the operations of intersection, union and complement on covering-based fuzzy rough sets are investigated. We first propose the concepts of intersection, union and complement of covering-based fuzzy rough sets.

Definition 4.1. Let $(U, ℂ)$ be a covering approximation space. For any $(\underline{C} (X), \bar{C} (X))$ , $(\underline{C} (Y), \bar{C} (Y)) \in \tilde{C}$ , the intersection and union are defined as follows: $\begin{matrix} (1) (\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y)) \\ = (\underline{C} (X) \cap \underline{C} (Y), \bar{C} (X) \cap \bar{C} (Y)); \\ (2) (\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y)) \\ = (\underline{C} (X) \cup \underline{C} (Y), \bar{C} (X) \cup \bar{C} (Y)) . \end{matrix}$

Definition 4.2. Let $(U, ℂ)$ be a covering approximation space. For any $(\underline{C} (X), \bar{C} (X)) \in \tilde{C}$ , the complement is defined as follows: $\sim (\underline{C} (X), \bar{C} (X)) = (\sim \bar{C} (X), \sim \underline{C} (X))$

Here, a question is raised: do all the covering-based fuzzy rough sets satisfy the operations of intersection, union and complement as defined above? The following will employ an example to illustrate the question.

Example 4.1. Let U = {x₁, x₂, x₃} , K₁ = {x₁, x₂}, K₂ = {x₂, x₃} , $ℂ = {K_{1}, K_{2}} .$ Clearly, $ℂ$ is a covering of U . For fuzzy sets X = (0.1, 0.2, 0.3) and Y = (0.3, 0.2, 0.1), we have $\underline{C} (X) \cap \underline{C} (X) = (0.1, 0.2, 0.1)$ and $\bar{C} (X) \cap \bar{C} (Y) = (0.2, 0.2, 0.2) .$ Clearly, there does not exist a fuzzy set V ∈ F (U) such that $(\underline{C} (V), \bar{C} (V)) = (\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y))$ .

Example 4.1 indicates that all the covering-based fuzzy rough sets do not meet the operation of intersection.

Let $(U, ℂ)$ be a monotone covering approximation space. If $ℂ$ satisfies the following condition (*).

For any x ∈ U, either $K_{x}^{min} = {x}$ or there exists y ∈ U such that $K_{x}^{min} = K_{y}^{min}$ . (*)

Thus, we can present the following proposition.

Proposition 4.1.Let $(U, ℂ)$ be a monotone covering approximation space and $ℂ$ satis fy the condition (*). For anyX, Y ∈ F (U), the following properties hold: $\begin{matrix} (1) \underline{C} (V) = \underline{C} (X) \cap \underline{C} (Y), \bar{C} (V) = \bar{C} (X) \cap \bar{C} (Y); \\ (2) \underline{C} (W) = \underline{C} (X) \cup \underline{C} (Y), \bar{C} (W) = \bar{C} (X) \cup \bar{C} (Y) . \end{matrix}$

Proof: (1) Suppose that $ℂ^{MIN} = {K_{1}, K_{2}, \dots, K_{m}} .$

For $K_{1} \in ℂ^{MIN}$ , choose x₁₁ ∈ K₁ randomly. According to Remark 2.1, $K_{1} = K_{x_{11}}^{min}$ can be obtained. Let $st (x_{11}, ℂ) = {K_{x_{11}}^{1}, K_{x_{11}}^{2}, \dots, K_{x_{11}}^{n}} .$ Since $ℂ$ is a monotone covering, without loss of generality, suppose that $x_{11} \in K_{x_{11}}^{min} = K_{x_{11}}^{1} \subset K_{x_{11}}^{2} \subset \dots \subset K_{x_{11}}^{n} .$

(1_{K
₁}) If $K_{x_{11}}^{1}$ has only one element, i.e., $K_{x_{11}}^{1} = {x_{11}}$ .

Denote $V (x_{11}) = (\underline{C} (X) \cap \underline{C} (Y)) (x_{11}) = (\bar{C} (X) \cap$ $\bar{C} (Y)) (x_{11})$ according to Proposition 3.5. If $K_{x_{11}}^{1}$ has atleast two elements, i.e., ${x_{11}} \subset K_{x_{11}}^{1}$ . According toProposition 3.6, we denote $V (x_{11}) = (\underline{C} (X) \cap \underline{C} (Y))$ (x₁₁) . For $\forall x \in K_{x_{11}}^{1} / {x_{11}},$ let $V (x) = (\bar{C} (X) \cap \bar{C} (Y))$ (x).

(2_{K
₁}) If $K_{x_{11}}^{2} / K_{x_{11}}^{1}$ has only one element, suppose that $K_{x_{11}}^{2} / K_{x_{11}}^{1} = {x_{12}}$ . According to proposition 3.5, we denote $V (x_{12}) = (\underline{C} (X) \cap \underline{C} (Y)) (x_{12}) = \bar{C} (X) \cap \bar{C} (Y) (x_{12})$ . If $K_{x_{11}}^{2} / K_{x_{11}}^{1}$ has at least two elements, according to Proposition 3.6, denote $V (x_{12}) = (\underline{C} (X) \cap \underline{C} (Y)) (x_{12})$ . For $\forall x \in K_{x_{11}}^{2} / (K_{x_{11}}^{1} \cup {x_{12}}),$ let $V (x) = (\bar{C} (X) \cap \bar{C} (Y)) (x)$ .

(3_{K
₁}) If $K_{x_{11}}^{3} / K_{x_{11}}^{2}$ has only one element, suppose that $K_{x_{11}}^{3} / K_{x_{11}}^{2} = {x_{13}}$ . According to Proposition 3.5, denote $V (x_{13}) = (\underline{C} (X) \cap \underline{C} (Y)) (x_{13}) = \bar{C} (X) \cap \bar{C} (Y) (x_{13})$ . If $K_{x_{11}}^{3} / K_{x_{11}}^{2}$ has at least two elements, according to Proposition 3.6, we denote $V (x_{13}) = (\underline{C} (X) \cap \underline{C} (Y)) (x_{13})$ . For $\forall x \in K_{x_{11}}^{3} / (K_{x_{11}}^{2} \cup {x_{13}}),$ let $V (x) = (\bar{C} (X) \cap \bar{C} (Y)) (x)$ .

⋯⋯

(n_{K
₁}) If $K_{x_{11}}^{n} / K_{x_{11}}^{n - 1}$ has only one element, suppose that $K_{x_{11}}^{n} / K_{x_{11}}^{n - 1} = {x_{1 n}}$ . According to Proposition 3.5, denote $V (x_{1 n}) = (\underline{C} (X) \cap \underline{C} (Y)) (x_{1 n}) = \bar{C} (X) \cap \bar{C} (Y) (x_{1 n})$ . If $K_{x_{11}}^{n} / K_{x_{11}}^{n - 1}$ has at least two elements, according to Proposition 3.6, denote $V (x_{1 n}) = (\underline{C} (X) \cap \underline{C} (Y)) (x_{1 n})$ . For $\forall x \in K_{x_{11}}^{n} / (K_{x_{11}}^{n - 1} \cup {x_{1 n}}),$ let $V (x) = (\bar{C} (X) \cap \bar{C} (Y)) (x)$ .

For $K_{i} \in ℂ^{MIN}, i = 2, 3, \dots, m$ , repeat the above steps from 1_{K
₁} to n_{K
₁}, respectively. Finally, a fuzzy set V ∈ F (U) is obtained such that $\underline{C} (V) = \underline{C} (X) \cap \underline{C} (Y)$ and $\bar{C} (V) = \bar{C} (X) \cap \bar{C} (Y)$ .

(2) The property can be proved according to (1).□

Remark 4.1. For any two covering-based fuzzy roughsets $(\underline{C} (X), \bar{C} (X))$ , $(\underline{C} (Y), \bar{C} (Y)) \in \tilde{C}$ , two fuzzy setsV, W ∈ F (U) can be obtained such that $(\underline{C} (V),$ $\bar{C} (V)) = (\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y))$ and $(\underline{C} (W),$ $\bar{C} (W)) = (\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y))$ .

Example 4.2. (Continued from Example 2.1) It is clear that $ℂ$ satisfies condition (*). For two fuzzy setsX = (0.3, 0.1, 0.2, 0.4, 0.5, 0.4, 0.3, 0.4) and Y = (0.2, 0.3, 0.4, 0.2, 0.6, 0.3, 0.2, 0.1), fuzzy sets V = (0.1, 0.5, 0.2, 0.5, 0.5, 0.1, 0.3, 0.1) and W = (0.2, 0.6, 0.2, 0.6, 0.6, 0.3, 0.4, 0.4) can be obtained such that $(\underline{C} (V), \bar{C} (V)) = (\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y))$ and $(\underline{C} (W),$ $\bar{C} (W)) = (\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y))$ .

Moreover, for Z = (0.2, 0.3, 0.4, 0.5, 0.4, 0.6, 0.1, 0.3), Let C = (0.2, 0.5, 0.4, 0.5, 0.5, 0.1, 0.6, 0.3), D = (0.2, 0.5, 0.2, 0.5, 0.4, 0.1, 0.4, 0.3), then $(\underline{C} (C),$ $\bar{C} (C)) = ((\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y))) \cup (\underline{C} (Z), \bar{C} (Z))$ and $(\underline{C} (D), \bar{C} (D)) = ((\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y))) \cap (\underline{C} (Z), \bar{C} (Z)) .$

4.2 Algebraic properties of covering-based fuzzy rough sets

In this section, the algebraic properties of covering-based fuzzy rough sets are investigated. Suppose that $(U, ℂ)$ is a monotone covering approximation space and that $ℂ$ satisfies condition (*). Then, we obtain the following conclusions.

Proposition 4.3. $(\tilde{C}, \cup, \cap, \sim)$ is an assignment lattice.

Proof: For any $(\underline{C} (X), \bar{C} (X))$ , $(\underline{C} (Y), \bar{C} (Y))$ and $(\underline{C} (Z), \bar{C} (Z)) \in \tilde{C}$ , we have $\begin{matrix} (1) (\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (X), \bar{C} (X)) \\ = (\underline{C} (X), \bar{C} (X)) (\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (X), \bar{C} (X)) \\ = (\underline{C} (X), \bar{C} (X)) \\ (2) (\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y)) \\ = (\underline{C} (Y), \bar{C} (Y)) \cap (\underline{C} (X), \bar{C} (X)) \\ (\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y)) \\ = (\underline{C} (Y), \bar{C} (Y)) \cup (\underline{C} (X), \bar{C} (X)) \\ (3) ((\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y))) \cap (\underline{C} (Z), \bar{C} (Z)) \\ = ((\underline{C} (X), \bar{C} (X)) \cap ((\underline{C} (Y), \bar{C} (Y)) \cap (\underline{C} (Z), \bar{C} (Z))) \\ ((\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y))) \cup (\underline{C} (Z), \bar{C} (Z)) \\ = ((\underline{C} (X), \bar{C} (X)) \cup ((\underline{C} (Y), \bar{C} (Y)) \cup (\underline{C} (Z), \bar{C} (Z))) \\ (4) ((\underline{C} (X), \bar{C} (X)) \cap ((\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y))) \\ = (\underline{C} (X), \bar{C} (X)) \\ ((\underline{C} (X), \bar{C} (X)) \cup ((\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y))) \\ = (\underline{C} (X), \bar{C} (X)) \\ (5) ((\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Y), \bar{C} (Y))) \cup (\underline{C} (Z), \bar{C} (Z)) \\ = ((\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Z), \bar{C} (Z))) \\ \cap ((\underline{C} (Y), \bar{C} (Y)) \cup (\underline{C} (Z), \bar{C} (Z))) \\ ((\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (Y), \bar{C} (Y))) \cap (\underline{C} (Z), \bar{C} (Z)) \\ = ((\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (Z), \bar{C} (Z))) \\ \cup ((\underline{C} (Y), \bar{C} (Y)) \cap (\underline{C} (Z), \bar{C} (Z))) . \end{matrix}$

Hence, $(\tilde{C}, \cup, \cap, \sim)$ is an assignment lattice.□

Let (φ, φ) =0, (U, U) =1, then the following conclusion is obtained.

Proposition 4.4. $(\tilde{C}, \cup, \cap, \sim)$ is a soft algebra.

Proof: (1) For any $(\underline{C} (X), \bar{C} (X))$ , $(\underline{C} (Y), \bar{C} (Y)) \in \tilde{C}$ , $\begin{matrix} (\underline{C} (X), \bar{C} (X)) \cup 0 = (\underline{C} (X), \bar{C} (X)) \\ (\underline{C} (X), \bar{C} (X)) \cap 0 = 0 \\ (\underline{C} (X), \bar{C} (X)) \cup 1 = 1 \\ (\underline{C} (X), \bar{C} (X)) \cap 1 = (\underline{C} (X), \bar{C} (X)) . \end{matrix}$

Thus, 0 and 1 are the minimal and maximal element of $(\tilde{C}, \cup, \cap, \sim)$ , respectively. $\begin{matrix} (2) \sim (\sim (\underline{C} (X), \bar{C} (X))) = \sim (\sim \bar{C} (X), \sim \underline{C} (X)) \\ = (\underline{C} (X), \bar{C} (X)) \\ (3) \sim ((\underline{C} (X), \bar{C} (X)) \cap (\underline{C} (X), \bar{C} (X))) \\ = \sim ((\underline{C} (X) \cap \underline{C} (Y), \bar{C} (X) \cap \bar{C} (Y))) \\ = (\sim (\bar{C} (X) \cap \bar{C} (Y)), \sim (\underline{C} (X) \cap \underline{C} (Y))) \\ = (\sim \bar{C} (X) \cup \sim \bar{C} (Y), \sim \underline{C} (X) \cup \sim \underline{C} (Y)) \\ = (\sim \bar{C} (X), \sim \underline{C} (X)) \cup (\sim \bar{C} (Y), \sim \underline{C} (Y)) \\ = \sim (\underline{C} (X), \bar{C} (X)) \cup \sim (\underline{C} (Y), \bar{C} (Y)) . \end{matrix}$

Similarly, $\begin{matrix} \sim ((\underline{C} (X), \bar{C} (X)) \cup (\underline{C} (X), \bar{C} (X))) \\ = \sim (\underline{C} (X), \bar{C} (X)) \cap \sim (\underline{C} (Y), \bar{C} (Y)) . \end{matrix}$

Hence, $(\tilde{C}, \cup, \cap, \sim)$ is a soft algebra.□

5 Conclusion

To easily deal with problems of uncertainty and imprecision, Xu, et al. proposed the multi-granulation fuzzy rough set model based on equivalence relations [23]. The model is a meaningful contribution toward the generalization of the classical rough set model.

It is well known that a multi-granulation rough set is a generalization of a Pawlak rough set. Covering-based rough sets are also an important generalization of classical rough sets. In this paper, we proposed the covering-based fuzzy rough set model and discussed its corresponding properties. Although many researchers have studied many properties of rough sets, the operations of intersection, union and complement on rough sets have yet to be investigated. In this paper, we proposed the concept of monotone covering and researched the operations of intersection, union and complement on covering-based fuzzy rough sets. Thus, the construction of the covering-based fuzzy rough set model is a meaningful generalization of rough set theory.

Footnotes

Acknowledgments

This work is supported by the Natural Science Foundation of China (No. 11161003, 11261006, 61472463 and 61402064) and the Science-Technology Foundation of the Education Department of Fujian Province, China (No. JA15281).

References

Radzikowska

Kerre

2002

A comparative study of fuzzy rough sets

Fuzzy Sets and Systems 126 137 155

Dubois

Prade

1992

Putting rough sets and fuzzy ests together

Intelligent Decision Support: Handbook of Applications and Advances of the Sets Theory Slowinski

203 232

Kluwer

Dordrecht

Chen

Wang

2007

A new approach to attribute reduction of consistent and inconsistent covering decision systems with covering rough sets

InformationSciences 177 3500 3518

Chen

Zhao

2010

Local reduction of decision system with fuzzy rough sets

Fuzzy Sets and Systems 161 1871 1883

Feng

Liu

Leoreanu-Fotea

Jun

2011

Soft sets and soft rough sets

Information Sciences 181 1125 1137

Feng

Davvaz

Ali

2010

Soft sets combined with fuzzy sets and rough sets; A tentative approach?

Soft Computing 14 899 911

Jeon

Kim

Jeong

2006

Rough sets attributes reduction based expert system in interlaced video sequences

IEEE Transactions on Consumer Electronics 52 1348 1355

Liu

2010

Closures and topological closures in quasi-discrete closure

Applied Mathematics Letters 23 772 776

Liu

Zhu

2008

The algebraic structures of generalized rough set theory

Information Sciences 178 4105 4113

10.

Liu

Sai

2009

A comparison of two types of rough sets induced by coverings

International Journal of Approximate Reasoning 50 521 528

11.

Pomkala

1987

Approximation operations in approximation space

Bulletin of the Polish Academy of Sciences 9–10 653 662

12.

Mei

2012

Knowledge reduction in real decision formal contexts

Information Sciences 189 191 207

13.

Qin

Gao

Pei

2007

On covering rough sets

The Second International Conference on Rough Sets and Knowledge Technology (RSKT 2007), Lecture Notes in Computer Science 4481 34 41

14.

Kondo

2006

On the structure of generalized rough sets

Information Sciences 176 589 600

15.

Kryaskiewics

1998

Rough set approach to incomplete information systems

Information Sciences 112 39 49

16.

Morsi

Yakout

1998

Axiomation for fuzzy rough sets

Fuzzy Sets and Systems 100 327 342

17.

Swiniarski

Skowron

2003

Rough set method in feature selection and recognition

Pattern Recognition Letter 24 833 849

18.

Yang

2010

Reduction about approximation spaces of covering generalized rough sets

International Journal of Approximate Reasoning 51 335 345

19.

Ananthanarayana

Narasimha

Subramanian

2003

Tree structure for efficient data mining usingrough sets

Pattern Recognition Letter 24 851 862

20.

Leung

2005

On characterizations of (I, T)-fuzzy rough approximation operators

Fuzzy Sets and Systems 154 76 102

21.

Zhang

2003

Generalized fuzzy rough sets

Information Sciences 151 263 282

22.

Wei

Liang

Qian

2012

A comparative study of rough sets for hybrid data

Information Sciences 190 1 16

23.

Wang

Luo

2014

Multi-granulation fuzzy rough sets

Journal of Intelligent and Fuzzy Systems 26 1323 1340

24.

Zakowski

1983

Approximations in the space (μ, π)

Demonstration Mathematica 16 761 769

25.

Zhu

2007

Basics concepts in covering-based rough sets

Third International Conference on Natural Computation 5 283 286

26.

Zhu

2007

Topological approaches to covering rough sets

Information Sciences 177 6 1499 1508

27.

Zhu

Wang

2007

On three types of covering rough sets

IEEE Transactions on Knowledge Data Engineering 19 8 1131 1144

28.

Zhu

2007

Generalized rough sets based on relations

Information Sciences 177 22 4997 5001

29.

Zhu

Wang

2003

Reduction and axiomatization of covering generalized rough sets

Information Sciences 152 217 230

30.

2011

Definable subsets in covering approximation spaces

International Journal Computing Mathematical Sciences 5 1 31 34

31.

Yao

1998

Generalized rough sets models

In: Knowledge Discovery Polkowski

Skowron

284 318

Physica-Verlag

Heidelberg

32.

Yao

2010

Three-way decisions with probabilistic rough sets

Information Sciences 180 3 341 353

33.

Yao

2012

Covering based rough set approximations

Information Sciences 200 1 91 107

34.

Bonikowski

Brynirski

Wybraniec

1998

Extensions and intentions in the rough set theory

Information Sciences 107 149 167

35.

Pawlak

1982

Rough sets

International Journal of Computer and Information Sciences 11 5 341 356