CRITERIA REDUCTION OF SET-VALUED ORDERED DECISION SYSTEM BASED ON APPROXIMATION QUALITY
|
|
- Rose Morton
- 6 years ago
- Views:
Transcription
1 International Journal of Innovative omputing, Information and ontrol II International c 2013 ISSN Volume 9, Number 6, June 2013 pp RITERI REDUTION OF SET-VLUED ORDERED DEISION SYSTEM BSED ON PPROXIMTION QULITY Zichun hen 1, Peng Shi 2,3, Penghui Liu 1 and Zheng Pei 1 1 School of Mathematics and omputer Engineering Xihua University hengdu , P. R. hina czclph@163.com 2 School of Engineering and Science Victoria University Melbourne, Vic 8001, ustralia Peng.Shi@vu.edu.au 3 School of Electrical and Electronic Engineering The University of delaide delaide, S 5005, ustralia Received March 2012; revised July 2012 bstract. The purpose of this paper is to further investigate criteria reduction in the set-valued ordered decision system, in which some objects may have more than one value for an attribute with preference-ordered domains. By considering the uncertainty of criteria values in such information system, we define a similarity dominance relation and give the dominance-based rough approximation of the upward and downward unions. To obtain criteria reduction from the inconsistent set-valued ordered decision system, the upward and downward approximation quality is proposed. Two types of criteria reduction are then formed for preserving the approximation quality. Furthermore, by the significance of the criterion, an algorithm is designed to compute a smallest reduction of a given set-valued ordered decision system. nd finally, a numerical example is employed to substantiate the conceptual arguments. Keywords: Set-valued ordered decision systems, Similarity dominance relation, Rough set, pproximation quality, riteria reduction 1. Introduction. Rough set theory [13], proposed by Pawlak, is an effective tool for data analysis. It can be used in information systems to describe the dependencies among attributes, evaluate the significance of attributes and derive decision rules. The Pawlak s rough set model is constructed based on equivalence relations which are viewed by many to be one of the main limitations [14, 15]. By relaxing equivalence relations to more general binary relation, the Pawlak s rough set can be extended to a more general model [1, 2, 8, 19, 23, 25, 27]. ttributes reduction is one of the important topics in rough set theory, its aim is to find a minimal attribute subset of the original data sets that is the most informative, and all other attributes can be deleted from the databases with the minimal information loss. Many types of attributes reduction based on the equivalence relation have been proposed. For instance, discernibility matrix [20], consistency of data [12], dependency of attributes [24] and information entropy [11] were employed to find reduction of an information system. In recent years, more attention has been paid to attribute reduction based on the more general binary relation and covering in rough set research [3, 8, 11, 16, 17, 22, 26, 27]. 2393
2 2394 Z. HEN, P. SHI, P. LIU ND Z. PEI lthough the classical rough set approach is a powerful tool for handling many problems, it is not able to deal with inconsistencies originating from the criteria, e.g., attributes with preference-ordered domains like product quality, market share, and debt ratio [7]. To solve this problem, Greco et al. have proposed an extension of Pawlak s rough set approach, which is called the Dominance-based Rough Set pproach (DRS) to take into account the ordering properties of criteria [5, 6, 7]. This innovation is mainly based on substitution of the equivalence relation by a similarity dominance relation which is reflexive and transitive. In recent years, many studies have been made in DRS [4, 10, 16, 17, 18, 21, 22, 26, 29]. Set-valued information systems are an important type of data table, and generalized of single-valued information systems. In such information systems, some objects may have more than one values which used to characterize uncertain information and missing information [28]. From the viewpoint of semantics, set-valued information systems can be summarized into two categories: disjunctive and conjunctive systems, and incomplete information systems with some unknown attribute values or partial known attribute values can be viewed as disjunctive set-valued information systems [8, 17]. In recent years, some problems of decision making in the context of set-valued information systems have been studied [8, 9, 17, 27]. For example, in [17], by introducing two dominance relations to the two types of set-valued information systems, Qian et al. investigated the problems of criteria reduction and decision rules extracted from these two types of information systems and decision tables. Nevertheless, the authors only considered the criteria reduction of the consistent set-valued ordered decision table, and the presented dominance relation in the disjunctive set-valued ordered information system was not transitive. Due to the rampant existence of the inconsistent systems in real life, new approach to knowledge reduction in the inconsistent set-valued ordered decision system has become a necessity. The purpose of this paper is to further study criteria reduction of the inconsistent set-valued ordered decision system interpreted disjunctively from the viewpoint of approximation quality. By introducing a similarity dominance relation in the set-valued ordered information system, we propose the concept of upward and downward approximation quality to characterize the relevance between criteria and decisions of set-valued ordered decision system (SODS), and define two types of criteria reduction by means of approximation quality. Furthermore, by the significance of the criterion, an algorithm is designed to obtain a smallest criteria reduction in inconsistent SODS. The rest of this paper is organized as follows: we first present some basic concepts of rough set theory and DRS in Section 2. In Section 3, we introduce a similarity dominance relation in the SODS, by which, a dominance-based rough set approach to SODS is established in Section 4. In Section 5, we define approximation quality to characterize the relevance between criteria and decision of the SODS, and propose the criteria reduction preserving the upward and downward approximation quality. n illustrative example is analyzed in Section 6 to show the feasibility of the conceptual arguments. Results are summarized in Section Some Basic oncepts of Rough Set Theory and DRS. n information system (IS) is a pair S = (U, T ), where U = {x 1,..., x n } is a nonempty finite set of objects and T = {a 1,..., a m } a nonempty finite set of attributes. With every subset of attributes T a binary relation Ind(), called the -indiscernibility relation, is defined by Ind() = {(x, y) U U : a, a(x) = a(y)}, where a(x) is the feature value of object x, then Ind() is an equivalence relation. By [x] we denote the equivalence class of x. For X U, the sets X = {x U : [x] B X} and X = {x U : [x] B X }
3 RITERI REDUTION OF SET-VLUED ORDERED DEISION SYSTEM 2395 are called the -lower and the -upper approximations of X, respectively. X is said to be definable if X = X; otherwise, X is a rough set. decision system (DS) is a pair S = (U, {d}), where d is called the decision attribute, and the elements in are called condition attributes. We say a is relatively dispensable in if P os (d) = P os {a} (d). Otherwise, a is said to be relatively indispensable in. Here P os (d) is the union of -lower approximation of all the equivalence classes of d, i.e., P os (d) = X U/d X. If every attribute in is relatively indispensable, we say that is relatively independent in S. subset is called a relative reduction in S if is relatively independent in S and P os (d) = P os (d). The collection of all relatively indispensable attributes in is called the relative core of S. The approximation quality of d with respect to is defined by γ (d) = P os (d) U The approximation quality is also called dependency, reflecting relevance between condition and decision. For multi-criteria decision analysis, each object in an IS can be seen as a sample decision, and each condition attribute is a criterion for that decision. criterion s domain of values is usually ordered according to the decision-maker s preferences. In general, an increasing preference and a decreasing preference are considered by a decision maker. In the following we only consider criteria with increasing preference. Definition 2.1. [22] n IS or DS is called a ordered information system (OIS) or ordered decision system (ODS) if all conduction attributes are criteria. Within the basic DRS, the notions of weak preference relation a and P -dominance relation D P are defined as follows. For a criterion a and any x, y U, x a y means that x is at least as good as (is weakly preferred to) y with respect to criterion a. The weak preference relation a is supposed to be a complete pre-order, i.e., complete, reflexive, transitive, and antisymmetric relation [4]. Moreover, for a subset of criteria P, the P - dominance relation D P is defined by D P = {(x, y) U U : a P, x a y}. When (x, y) D P, we say that x P -dominates y, and that y is P -dominated by x. Given the dominance relation D P, the P -dominating set and P -dominated set of x are defined as D + P (x) = {y U : (y, x) D P } and D P (x) = {y U : (x, y) D P }, respectively. In addition, it is assumed that the decision attribute d makes a partition of U into a finite number of classes D = {D 1, D 2,..., D r }, which are ordered, that is, for all i, j r, if i j, then the objects from D i are preferred to the objects from D j. In order to reflect the total order and dominance relations, the upward and downward unions are defined respectively as D i = j i D j and D i = j i D j. Then, we have D 1 = Dr = U, Dr = D r, D 1 = D 1, D i = U D i 1, D i = U D i+1, (i 1, r) (2) The statement x D i means x belongs to at least class D i, whereas x D i means x belongs to at most class D i [7]. Furthermore, the P -lower and P -upper approximations of D i and D i are defined as P (D i ) = {x U : D+ P (x) D i }, P (D i ) = {x U : D P (x) D i } (3) P (D i ) = {x U : D P (x) D i }, P (D i ) = {x U : D+ P (x) D i } (4) 3. Similarity Dominance Relation. In many practical issues, it may happen that some of the attribute values for an object are set-valued, which are always used to characterize uncertain information and missing information in information systems. (1)
4 2396 Z. HEN, P. SHI, P. LIU ND Z. PEI Definition 3.1. [17] n information system S = (U, T ) is called a set-valued information system (SIS), if a T, x U, a(x) 2 Va { }, where V a is the domain of a attribute a. decision system S = (U, d) is called a set-valued decision system (SDS), if (U, ) is a SIS, (U, d) is a single-valued information system. Table 1 illustrates an SDS, where U = {x 1, x 2,..., x 8 }, = {a 1, a 2, a 3, a 4 } is the set of condition attribute, d is a decision attribute. Table 1. set-valued ordered decision information x 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 a 1 {1} {1} {1,2} {2} {3} {1} {1,2} {2,3} a 2 {1} {1,2} {2} {2} {1} {1} {2} {2} a 3 {1,2} {2,3} {3} {2,3} {1} {1,2} {2} {1} a 4 {2} {1} {1,2} {1} {1,2} {1,2} {1} {1,2} d From the viewpoint of semantic interpretation, the SIS can be classified into two categories: disjunctive and conjunctive system [8]. In this paper, we focus on the set-valued ordered decision information (SODS) interpreted disjunctively, also denote S = (U, d), where all conduction attributes are criteria and a, x U, each value in the subset a(x) is equally possible. In a set-valued ordered information system, the objects may have more than one evaluations with respect to the condition criteria. The dominance relations between the objects cannot be determined with certainty. Thus, we define a weak preference relation a on V a based on one possibility x dominates y with respect to criterion a as follows: x a y if and only if v y a a(y), v x a a(x) such that v x a v y a If x a y, then for max a(y), there v x a a(x) such that v x a max a(y), so max a(x) v x a max a(y). onversely, suppose max a(x) max a(y), then v y a a(y), there max a(x) a(x) such that max a(x) max a(y) v y a, so x a y. Thus, x a y max a(x) max a(y) (5) ccording to Equation (5), It easily obtain that the weak preference relation a is a completely pre-ordered on V a. Definition 3.2. Let S = (U, d) be a SODS,, we define a dominance relation R as follows: R = {(x, y) U 2 : a, x a y}. (6) Lemma 3.1. Let S = (U, d) be a SODS,, we have that (1) R is a similarity dominance relation, i.e., R is reflexive and transitive; (2) R = a R a, where Ra = R {a} ; (3) if B, then R R R B. Remark 3.1. In [17], Qian et al. defined a binary dominance relation in the disjunctive set-valued ordered information system as follows: R = {(x, y) U 2 : a, max a(x) min a(y)}. However, R is only reflexive, not transitive, i.e., it is not pre-ordered. By dominance relation R, we can obtain the -dominating set and the -dominated of x as follows: R + (x) = {y U : (y, x) R } = {y U : a, max a(y) max a(x)} (7) (x) = {y U : (x, y) R } = {y U : a, max a(x) max a(y)} (8)
5 RITERI REDUTION OF SET-VLUED ORDERED DEISION SYSTEM 2397 Lemma 3.2. Let S = (U, d) be a SODS,, B, x, y U, then we have that (1) B R + (x) R + B (x), R (x) R B (x); (2) y R + (x) R + (y) R + (x), y R (x) R (y) R (x); (3) R + (x) = {R + (y) : y R + (x)}, R (x) = { (y) : y R (x)}. Example 3.1. Let us compute the -dominating set and the -dominated based on dominance relation R. In Table 1, where = {a 1, a 2, a 3, a 4 }. By R + (x) = {y U : a, max a(y) max a(x)} and R (x) = {y U : a, max a(y) max a(x)}, one can get that R + (x 1) = {x 1, x 3, x 6 }, R + (x 2) = {x 2, x 3, x 4 }, R + (x 3) = {x 3 }, R + = {x 3, x 4 }, R + (x 5) = {x 5, x 8 }, R + (x 6) = {x 1, x 3, x 6 }, R + (x 7) = {x 3, x 4, x 7 }, R + (x 8) = {x 8 }. (x 1) = {x 1, x 6 }, (x 2) = {x 2 }, (x 3) = {x 1, x 2, x 3, x 4, x 6, x 7 }, (x 4) = {x 2, x 4, x 7 }, (x 5) = {x 5 }, (x 6) = {x 1, x 6 }, (x 7) = {x 7 }, (x 8) = {x 5, x 8 }. 4. Dominance-Based Rough pproximations. In this section, we will investigate dominance-based rough approximations of the upward and downward unions based on the similarity dominance relation and gives its some important properties. Let S = (U, {d}) be a given SODS, D = {D 1, D 2,, D r } be ordered decision classes induced by d, D = {D 1, D 2,, D r } and D = {D 1, D 2,, D r } be the upward unions and downward unions classes, respectively. Definition 4.1. Let S = (U, {d}) be a SODS,. Then, D i D, D i D, the -lower approximations and -upper approximations of D i and D i are defined as follows: (D i ) = {R + (x) : R + (x) D i }, (D i ) = {R + (x) : x D i } (9) (D i ) = { (x) : R (x) D i }, (D i ) = { (x) : x D i } (10) The lower and upper approximations of the upward and downward unions have the following important properties. Theorem 4.1. Let S = (U, {d}) be a SODS,, B, D i D, D i D, we have that (D i ) = U (D i 1 ), (i 2), (D i ) = U (D i+1 ), (i r 1) (11) (D 1 ) = (D 1 ) = U, (Dr ) = (Dr ) = U (12) (D i ) D i (D i ), (D i ) D i (D i ) (13) (D i ) = {x U : R + (x) D i } (14) (D i ) = {x U : R (x) D i } (15) (D i ) = {x U : R (x) D i } (16) (D i ) = {x U : R + (x) D i } (17) B B(D i ) (D i ), (D i ) B(D i ) (18) B B(D i ) (D i ), (D i ) B(D i ) (19) Proof: These proof can be concluded similarly in [10, 26]. Definition 4.2. Let S = (U, {d}) be a SODS, and denote R d = {(x, y) U 2 : d(x) d(y)}. If R R d, then S is called consistent; otherwise it is called inconsistent.
6 2398 Z. HEN, P. SHI, P. LIU ND Z. PEI ccording to Definition 4.2, in a consistent SODS, if x is better than y in terms of criterion a, then x s decision should not be worse than y s. Theorem 4.2. Let S = (U, {d}) be a SODS, then the following statements are equivalent: (1) S = (U, {d}, V, f) is consistent. (2) x U, R + (x) D j, where j = d(x). (3) i r, (D i ) = D i = (D i ). (4) x U, (x) D j, where j = d(x). (5) i r, (D i ) = D i = (D i ). Proof: (1) (2): x U, if y R + (x), then (y, x) R. Since S is consistent, so we have (y, x) R d, it follows that d(y) d(x) = j. Therefore, R + (x) D j holds. (2) (1): (y, x) R, i.e., y R + (x), by (2) we obtain y R + (x) D j, where j = d(x). Hence, d(y) j = d(x), that is, (y, x) R d. It follows that S is consistent. (2) (3): i r, if y (D i ), then by Definition 4.1, there exist x D i such that y R + (x). By (2) we obtain y D j, where j = d(x), this indicates d(y) d(x). However, x D i, d(x) i, so d(y) i, it follows that y D i ; hence, (D i ) D i. On the other hand, D i (D i ) by Equation (13), so we have D i = (D i ). Similarly, we can prove (D i ) = D i. (3) (2): x U. Since x D j, where j = d(x), so we have x (D j ) by (3). Thus, by Definition 4.1, we can obtain R + (x) D j, where j = d(x). The proofs of the rest (1) (4) (5) are similar to the proof of (1) (2) (3). 5. riteria Reduction in SODS. ttribute reduction is one of major topics in the rough set approach. By the method, superfluous condition attributes are removed so that we may find condition attributes related to the decision attribute. Through using discernibility matrices, Qian et al. have proposed criteria reduction to the consistent set-valued ordered decision system [17]. Due to the rampant existence of the inconsistent systems in real life, new approach to criteria reduction in the inconsistent set-valued ordered decision system has become a necessity. In this section, we will introduce approximation quality to characterize the relevance between condition criteria and decision of the SODS, and propose the criteria reduction preserving the upward and downward approximation quality. Definition 5.1. Let S = (U, {d}) be a SODS,, and D = {D 1, D 2,, D r } be ordered decision classes induced by d, denote by γ (d) = 1 r (D i ) r 1 D i ; (20) i=2 γ (d) = 1 r 1 r 1 i=1 (D i ) D i (21) then γ (d) and γ (d) are called upward and downward approximation quality of the subset of condition criteria relative to the decision d, respectively. Obviously, 0 γ (d) 1 and 0 γ (d) 1. Remark 5.1. It is noticed that (D 1 ) = (Dr ) = U for each, we cannot obtain any information about approximation quality from the lower of D 1 and Dr. Thus, the
7 RITERI REDUTION OF SET-VLUED ORDERED DEISION SYSTEM 2399 upward approximation quality is defined based on D i approximation quality based on D i (1 i r 1). (2 i r) and the downward Theorem 5.1. Let S = (U, {d}) be a SODS. If B, then we have γ B (d) γ (d) and γ B (d) γ (d). Proof: Since B, by Equations (18) and (19), we have B(D i ) (D i ) and B(D i ) (D i ) for 1 i r. Hence, γ B (d) γ (d), γ B (d) γ (d). By Theorem 5.1, a, we have γ {a} (d) γ (d) and γ {a} (d) γ (d). Theorem 5.2. Let S = (U, {d}) be a SODS, then we have that (1) S is a consistent SODS γ (d) = 1; (2) S is a consistent SODS γ (d) = 1. Proof: (1) If S is a consistent SODS, then for each D i, we have (D i ) = D i by Theorem 4.2. Hence, γ (d) = 1. onversely, suppose that γ (d) = 1 and S is a inconsistent SODS, then there is D j (2 j r) such that (D j ) D j by Theorem 4.2, i.e., (D j ) < D j. So we have γ (d) < 1, which is a contradiction. Hence, S is a consistent. (2) The proof of (2) is similar to the proof of (1). Definition 5.2. Let S = (U, {d}) be a SODS. For a, if γ {a} (d) = γ (d), then we say a is upward dispensable, otherwise a is upward indispensable and γ (d) > γ {a} (d). The collection of all the upward indispensable elements is called the upward core, denoted by ore (S). Similarly, if γ {a} (d) = γ (d), then we say a is downward dispensable, otherwise a is downward indispensable and γ (d) > γ {a} (d). The collection of all the downward indispensable elements is called the downward core, denoted by ore (S). Definition 5.3. Let S = (U, {d}) be a SODS,. We say is an upward reduction if (1) γ (d) = γ (d); (2) B, γ (d) > γ B (d). orrespondingly, we can define the downward reduction as follows. Definition 5.4. Let S = (U, {d}) be a SODS,. We say is an downward reduction if (1) γ (d) = γ (d); (2) B, γ (d) > γ B (d). Theorem 5.3. Let S = (U, {d}) be a SODS,, then we have that (1) if is an upward reduction, then ore (S) ; (2) if is an downward reduction, then ore (S). Proof: (1) a ore (S), i.e., γ (d) > γ {a} (d). If a / then {a}. Since is an upward reduction, we have γ (d) γ {a} (d) γ (d) = γ (d) by Definition 5.3, that is γ (d) = γ {a} (d), which is a contradiction. Hence, ore (S). (2) The proof of (2) is similar to the proof of (1). Theorem 5.4. Let S = (U, {d}) be a SODS,, then we have that (1) is an upward reduction if and only if i (2 i r), (D i ) = (D i ), and B, j (2 j r), B(D j ) (D j );
8 2400 Z. HEN, P. SHI, P. LIU ND Z. PEI (2) is an downward reduction if and only if i (2 i r), (D i ) = (D i B, j (2 j r), B(D j ) (D j ). ), and Proof: (1) Let is an upward reduction. Since (D i ) (D i ) for all i r by Equation (18), if there is i 0 such that (D i 0 ) (D i 0 ), that is (D i 0 ) < (D i 0 ), then we have γ (D) < γ (D) by Theorem 5.1, which is a contradiction. Hence, i (2 i r), (D i ) = (D i ). Furthermore, suppose that there is a B such that i (2 i r), B(D i ) = (D i ), then we have γ (d) = γ B (d), which is a also contradiction to γ (d) > γ B (d). onversely, It is straightforward by Definition 4.1. (2) The proof of (2) is similar to the proof of (1). Theorem 5.4 shows that an upward/downward reduction is a minimal condition criterion set preserving the lower approximations of all the upper/downward unions. Lemma 5.1. Let S = (U, {d}) be a SODS,, then we have that (1) i r, (D i ) = (D i ) (D i ) = (D i ); (2) i r, (D i ) = (D i ) (D i ) = (D i) ). Proof: We can easily derive these conclusions according to Equations (11) and (12). Theorem 5.5. Let S = (U, {d}) be a SODS,, then we have that (1) is an upward reduction if and only if i (2 i r), (D i ) = (D i ), and B, j (2 j r), B(D j ) (D j ); (2) is an downward reduction if and only if i (2 i r), (D i ) = (D i ), and B, j (2 j r), B(D j ) (D j ). Proof: These conclusions are straightforward according to Theorem 5.4 and Lemma 5.1. Theorem 5.5 shows that an upward/downward reduction is also a minimal condition criterion set preserving the upper approximations of all the downward/upper unions. Definition 5.5. Let S = (U, {d}) be a SODS, a, denote by Sig (a,, d) = γ (d) γ {a} (d) (22) Sig (a,, d) = γ (d) γ {a} (d) (23) then we say Sig (a,, d) and Sig (a,, d) are the upward and downward significance of a to d, respectively. The upward/downward significance of a to d characterize the capacity of the upward/downward approximation quality when is deleted from. By Definitions 5.2 and 5.5, we have the following result. Theorem 5.6. Let S = (U, {d}) be a SODS, then we have that (1) ore (S) = {a : Sig (a,, d) > 0}; (2) ore (S) = {a : Sig (a,, d) > 0}. Definition 5.6. Let S = (U, {d}) be a SODS,, a, a /, denote by Sig (a, d) = γ {a} (d) γ (d) (24) Sig (a, d) = γ {a} (d) γ (d) (25) then we say Sig (a, d) and Sig (a, d) are the relative upward and downward significance of criterion a to, respectively.
9 RITERI REDUTION OF SET-VLUED ORDERED DEISION SYSTEM 2401 The relative upward/downward significance of criterion a to can be used to measure the capacity of the upward/downward approximation quality when a is added to. The larger the value of Sig (a, d)/sig (a, d) is, the more important a to is. Now we can use the upward/downward significance and relative upward/downward significance of a criterion as heuristic knowledge to design an algorithm to compute all the relative reduction. Firstly, by Theorem 5.6, we can calculate ore (S) and ore (S). Secondly, since the upward/downward core of criteria is a subset of all upward/downward reduction by Theorem 5.3, we can take the ore (S)/ore (S) for original upward/downward reduction set and extend it one by one. ccording to the relative upward/downward significance of criterion, we can select the criterion with the biggest relative upward/downward significance to the upward/downward core and add it to the upward/downward core until its upward/downward approximation quality relative to d is equal to the upward/downward approximation quality of to d. lgorithm 1. Search of the upward core and the upward reduction of a given SODS. Input: a given SODS S = (U, {d}) Output: a upward reduction of the ordered decision table Step 1: ore (S) Step 2: For each a i, compute Sig (a i,, d). If Sig (c i, ) > 0 then ore (S) {a i } end end Step 3: if γ (d) = ore (S) γ (d) then Step 4: return ore (S) // Now, ore (S) is the smallest upward reduction Step 5: else ore (S) Red Step 6: for each a j { ore (S)}, compute Sig (a ore (S) j, d) end Step 7: if Sig (a ore (S) k, d) = max aj { ore (S)} Sig (a ore (S) j, d) Red {a k } Red, go to Step 6 else return Red Step 8: end Let = n and U = m, the time complexity of lgorithm 1 is O(n 3 m 2 ). Similarly, we can also develop an algorithm of an downward reduction. In the following, we use an example to illustrate feasibility of the algorithm. Example 5.1. (ontinued from Example 3.1). We calculate the smallest upward reduction and core by lgorithm 1 in Table 1. From Table 1, we have that D 1 = {x 1, x 6 }, D 2 = {x 2, x 3, x 4, x 7 }, D 3 = {x 5, x 8 }. D 2 = {x 2, x 3, x 4, x 5, x 7, x 8 }, D 3 = {x 5, x 8 }. First, we calculate the upward significance of each criterion. Put = {a 1, a 2, a 3, a 4 }, 1 = {a 2, a 3, a 4 }, 2 = {a 1, a 3, a 4 }, 3 = {a 1, a 2, a 4 }, 4 = {a 1, a 2, a 3 }. By Definition 5.1, we can obtain that γ (d) = 1, γ 1 (d) = 5, 12 γ 2 (d) = 1,γ 3 (d) = 1, γ 4 (d) = 1. Then, we can calculate the upward significance of each criterion by Definition 5.5 as follows: Sig (a 1,, d) = 7, 12 Sig (a 2,, d) = Sig (a 3,, d) = Sig (a 4,, d) = 0. Hence, ore (S) = {a 1 }. Since γ 11 (d) =, ore (S) 12 γ (d) = 1 and γ (d) ore (S) γ (d), we calculate the relative upward significance of other criteria to the upward core. Put B 1 = γ ore (S) = {a 1 }, for B 1 = {a 2, a 3, a 4 }, we have Sig B 1 (a 2, d) = 1, 12 Sig B 1 (a 3, d) = 1, 12 Sig B 1 (a 4, d) = 0 and γ {a 1,a 2 } (d) = γ (d), γ {a 1,a 3 } (d) = γ (d). So
10 2402 Z. HEN, P. SHI, P. LIU ND Z. PEI {a 1, a 2 } and {a 1, a 3 } are the set of the upward reduction in Table 1. Similarly, we can obtain {a 1, a 2 } and {a 1, a 3 } are also the set of the downward reduction in Table Numerical Example. In many practical decision-making issues, set-valued information systems and set-valued decision tables have very wide applications, which can be used in intelligent decision-making and knowledge discovery from information systems with uncertain information and set-valued information [17]. decision-maker may need to adopt a better one from some possible projects or find some directions from existing successful projects before making a decision. The purpose of this section is to illustrate how to obtain criteria reduction from the inconsistent SODS by using the approaches proposed in this paper. Let us consider a practical decision problem of a summary of reviewer s reports for papers submitted to a journal. Suppose U = {x 1,..., x 10 } be a set of ten evaluated papers, = {a 1, a 2, a 3, a 4 } be a set of condition criteria, d is the decision attribute, where a 1 means originality, a 2 means presentation, a 3 means technical soundness, a 4 means motivation, d means overall evaluation. The values of these criteria are V a1 = V a2 = V a3 = V a4 = {0(poor), 1(fair), 2(good)}, V d = {1(reject), 2(revision), 3(accept)}. ssume that we have three specialists to evaluate the criterion values for these papers, and the evaluations given by these specialists are of the same importance, then it is possible that their evaluation results in the same criterion are not the same as one another. If we want to combine these evaluations together without losing information, we should union the evaluations given by every specialist for every attribute value. Furthermore, if we suppose here that poor < fair < good, reject < revision < ccept are two preference, then this problem can be stated by a set-valued ordered decision system as shown in Table 2. Table 2. summary of reviewer s reports for ten papers x 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 x 9 x 10 a 1 {1} {0,1} {0} {0} {2} {0,2} {1} {0} {1} {1} a 2 {1,2} {0} {0,1} {1} {0} {0} {1} {0} {1} {0,1} a 3 {1} {2} {1,2} {1} {1} {1} {0,2} {2} {0,1} {1} a 4 {2} {0} {0,2} {0,2} {1} {1} {2} {0,1} {2} {2} d From Table 2, Equations (7) and (8), we have that R + (x 1) = {x 1 }, R + (x 2) = {x 2, x 7 }, R + (x 3) = {x 3, x 7 }, R + (x 4) = {x 1, x 3, x 4, x 7, x 9, x 10 }, R + (x 5) = R + (x 6) = {x 5, x 6 }, R + (x 7) = {x 7 }, R + (x 8) = {x 3, x 7, x 8 }, R + (x 9) = R + (x 10) = {x 1, x 7, x 9, x 10 }; (x 1) = {x 1, x 4, x 9, x 10 }, (x 2) = {x 2 }, (x 3) = {x 3, x 4, x 8 }, (x 4) = {x 4 }, (x 5) = (x 6) = {x 5, x 6 }, (x 7) = {x 2, x 3, x 4, x 7, x 8, x 9, x 10 }, (x 8) = {x 8 }, (x 9) = (x 10) = {x 4, x 9, x 10 }. The decision attribute d makes a partition of U into a finite number of classes such that D = {D 1, D 2, D 3 }, where D 1 = {reject} = {x 2, x 3, x 8 }, D 2 = {revision} = {x 4, x 5, x 6, x 10 }, D 3 = {accept} = {x 1, x 7, x 9 }. Thus, D 2 = D 2 D3 = {x 1, x 4, x 5, x 6, x 7, x 9, x 10 }, D 3 = D 3 = {x 1, x 7, x 9 }, D 1 = D 1 = {x 2, x 3, x 8 }, D 2 = D 1 D2 = {x 2, x 3, x 4, x 5, x 6, x 8, x 10 }. It must be noticed that (x 3, x 4 ) R and (x 3, x 4 ) / R d, then Table 2 is inconsistent.
11 RITERI REDUTION OF SET-VLUED ORDERED DEISION SYSTEM 2403 By Definitions 4.1 and 5.1, we have (D 2 ) = {x 1, x 5, x 6, x 7, x 9, x 10 }, (D 3 ) = {x 1, x 7 }; (D 1 ) = {x 2, x 8 }, (D 2 ) = {x 2, x 3, x 4, x 5, x 6, x 8 }; γ 16 (d) =, 21 γ 16 (d) =. 21 Based on Definition 5.5, by computation, we have Sig (a 1,, d) = 11, 21 Sig (a 2,, d) = Sig (a 3,, d) = 1, 6 Sig (a 4,, d) = 0. Hence, ore (S) = {a 1, a 2, a 3 }. Since γ 16 (d) = = ore (S) 21 γ (d), By lgorithm 1, {a 1, a 2, a 3 }, i.e., {originality, presentation, technical soundness} is the upward reduction of Table 2. Similarly, it is not difficult to obtain {a 1, a 2 } and {a 1, a 3 } are the downward reduction of Table onclusions. Set-valued information systems are an important type of data table, and have very wide applications in many practical decision-making issues. In this paper, we have investigated criteria reduction in the inconsistent set-valued ordered decision system. By considering the uncertainty of criteria values in the set-valued ordered information system, we defined a similarity dominance relation. To obtain criteria reduction from the inconsistent set-valued ordered decision system, the upward and downward approximation quality are proposed and two types of criteria reduction are defined. nd finally, by the significance of the criterion, an algorithm is designed to compute a smallest reduction of a given set-valued ordered decision system. In further research, we will consider to extract and simplify the dominance decision rules in the inconsistent set-valued ordered decision system. cknowledgment. This work is partially supported by the research fund of Sichuan key laboratory of intelligent network information processing (SGXZD ), the National Natural Science Foundation ( , ), Sichuan Key Technology Research and Development Program (2012GZ0019, 2011FZ0051), the research fund of education department of Sichuan province (10Z058) and Supported by the open research fund of key laboratory of intelligent network information processing, Xihua University (SZJJ ). REFERENES [1] L. n, Z. hen and L. Tong, Rough set analysis of interval-valued information systems, II Express Letters, vol.5, no.8(b), pp , [2] R. Slowinski and D. Vanderpooten, generalized definition of rough approximations based on similarity, IEEE Trans. Data Knowl. Eng., vol.2, pp , [3] D. G. hen,. Z. Wang and Q. H. Hu, new approach to attribute reduction of consistent and inconsistent covering decision systems with covering rough sets, Information Sciences, vol.177, pp , [4] K. Dembcznski, S. Greco and R. Slowinski, Rough set approach to multiple criteria classification with imprecise evaluations and assignments, European Journal of Operational Research, vol.198, pp , [5] S. Greco, B. Matarazzo and R. Slowinski, Theory and methodology rough approximation of a preference relation by dominance relations, European Journal of Operational Research, vol.117, pp.63-83, [6] S. Greco, B. Matarazzo and R. Slowinski, Rough sets theory for multi-criteria decision analysis, European Journal of Operational Research, vol.129, pp.1-47, [7] S. Greco, B. Matarazzo and R. Slowinski, Rough approximation by dominance relation, International Journal of Intelligent Systems, vol.17, pp , [8] Y. Y. Guan and H. K. Wang, Set-valued information systems, Information Sciences, vol.176, pp , 2006 [9] Y. Leung, W. Z. Wu and W. X. Zhang, Knowledge acquisition in incomplete information systems: rough set approach, European Journal of Operational Research, vol.168, pp , 2006.
12 2404 Z. HEN, P. SHI, P. LIU ND Z. PEI [10] M. Inuiguchi, Y. Yoshioka and Y. Kusunoki, Variable-precision dominance-based rough set approach and attribute reduction, International Journal of pproximate Reasoning, [11] F. Li and Y. Q. Yin, pproaches to knowledge reduction of covering decision systems based on information theory, Information Sciences, vol.179, pp , [12] J. S. Mi, W. Z. Wu and W. X. Zhang, pproaches to knowledge reduction based on variable precision rough set model, Information Sciences, vol.159, pp , [13] Z. Pawlak, Rough Sets: Theoretical spects of Reasoning about Data, Kluwer cademic Publishers, Boston, [14] Z. Pawlak, Rough sets: Some extensions, Information Sciences, vol.177, pp.28-40, [15] Z. Pawlak and. Skowron, Rough sets and boolean reasoning, Information Sciences, vol.177, pp.41-73, [16] Y. Qian, J. Liang and. Dang, Interval ordered information systems, omputers and Mathematics with pplications, vol.56, pp , [17] Y. Qian,. Dang, J. Liang and D. Tang, Set-valued ordered information systems, Information Sciences, vol.179, pp , [18] J. J. H. Liou and G. H. Tzeng, dominance-based rough set approach to customer behavior in the airline market, Information Sciences, vol.180, pp , [19]. Skowron and J. Stepaniuk, Tolerance approximation spaces, Fundamenta Information, vol.27, pp , [20]. Skowron and. Rauszer, The discernibility matrices and functions in information systems, in Intelligent Decision Support, Handbook of pplications and dvances of the Rough Sets Theory, R. Slowinsk (ed.), Kluwer cademic Publishers, [21] R. Susmaga, R. Slowinski, S. Greco and B. Matarazzo, Generation of reducts and rules in multiattribute and multi-criteria classification, ontrol and ybernetics Series, vol.29, no.4, pp , [22] M. W. Shao and W. X. Zhang, Dominance relation and rules in an incomplete ordered information system, International Journal of Intelligent Systems, vol.20, pp.13-27, [23] K. Watjanapong and S. Yan, The characters recognition based on tolerant rough sets with fuzzy c-mean, II Express Letters, vol.3, no.4, pp , [24] J. Wang and D. Q. Miao, nalysis on attribute reduction strategies of rough set, J. omput. Sci. Technol, vol.13, pp , [25] Y. Y. Yao, Relational interpretations of neighborhood operators and rough set approximation operators, Information Sciences, vol.101, pp , [26] X. Yang, J. Yang,. Wu and D. Yu, Dominance-based rough set approach and knowledge reductions in incomplete ordered information system, Information Sciences, vol.178, pp , [27] W. X. Zhang, Incomplete information system and its optimal selections, omputers and Mathematics with pplications, vol.48, no.5, pp , [28] W. X. Zhang, Y. Leung and W. Z. Wu, Information Systems and Knowledge Discovery, Science Press, Beijing, hina, [29] L. Y. Zhai, L. P. Khoo and Z. W. Zhong, dominance-based rough set approach to Kansei Engineering in product development, Expert Systems with pplications, vol.36, pp , 2009.
Research Article Decision Rules Acquisition for Inconsistent Disjunctive Set-Valued Ordered Decision Information Systems
Mathematical Problems in Engineering Volume 2015, rticle ID 936340, 8 pages http://dx.doi.org/10.1155/2015/936340 Research rticle Decision Rules cquisition for Inconsistent Disjunctive Set-Valued Ordered
More informationARPN Journal of Science and Technology All rights reserved.
Rule Induction Based On Boundary Region Partition Reduction with Stards Comparisons Du Weifeng Min Xiao School of Mathematics Physics Information Engineering Jiaxing University Jiaxing 34 China ABSTRACT
More informationComputers and Mathematics with Applications
Computers and Mathematics with Applications 59 (2010) 431 436 Contents lists available at ScienceDirect Computers and Mathematics with Applications journal homepage: www.elsevier.com/locate/camwa A short
More informationMETRIC BASED ATTRIBUTE REDUCTION IN DYNAMIC DECISION TABLES
Annales Univ. Sci. Budapest., Sect. Comp. 42 2014 157 172 METRIC BASED ATTRIBUTE REDUCTION IN DYNAMIC DECISION TABLES János Demetrovics Budapest, Hungary Vu Duc Thi Ha Noi, Viet Nam Nguyen Long Giang Ha
More informationA Scientometrics Study of Rough Sets in Three Decades
A Scientometrics Study of Rough Sets in Three Decades JingTao Yao and Yan Zhang Department of Computer Science University of Regina [jtyao, zhang83y]@cs.uregina.ca Oct. 8, 2013 J. T. Yao & Y. Zhang A Scientometrics
More informationInterpreting Low and High Order Rules: A Granular Computing Approach
Interpreting Low and High Order Rules: A Granular Computing Approach Yiyu Yao, Bing Zhou and Yaohua Chen Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 E-mail:
More informationHierarchical Structures on Multigranulation Spaces
Yang XB, Qian YH, Yang JY. Hierarchical structures on multigranulation spaces. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 27(6): 1169 1183 Nov. 2012. DOI 10.1007/s11390-012-1294-0 Hierarchical Structures
More informationMinimal Attribute Space Bias for Attribute Reduction
Minimal Attribute Space Bias for Attribute Reduction Fan Min, Xianghui Du, Hang Qiu, and Qihe Liu School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu
More informationOn Improving the k-means Algorithm to Classify Unclassified Patterns
On Improving the k-means Algorithm to Classify Unclassified Patterns Mohamed M. Rizk 1, Safar Mohamed Safar Alghamdi 2 1 Mathematics & Statistics Department, Faculty of Science, Taif University, Taif,
More informationROUGH set methodology has been witnessed great success
IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 14, NO. 2, APRIL 2006 191 Fuzzy Probabilistic Approximation Spaces and Their Information Measures Qinghua Hu, Daren Yu, Zongxia Xie, and Jinfu Liu Abstract Rough
More information2 WANG Jue, CUI Jia et al. Vol.16 no", the discernibility matrix is only a new kind of learning method. Otherwise, we have to provide the specificatio
Vol.16 No.1 J. Comput. Sci. & Technol. Jan. 2001 Investigation on AQ11, ID3 and the Principle of Discernibility Matrix WANG Jue (Ξ ±), CUI Jia ( ) and ZHAO Kai (Π Λ) Institute of Automation, The Chinese
More informationThe Fourth International Conference on Innovative Computing, Information and Control
The Fourth International Conference on Innovative Computing, Information and Control December 7-9, 2009, Kaohsiung, Taiwan http://bit.kuas.edu.tw/~icic09 Dear Prof. Yann-Chang Huang, Thank you for your
More informationQuantization of Rough Set Based Attribute Reduction
A Journal of Software Engineering and Applications, 0, 5, 7 doi:46/sea05b0 Published Online Decemer 0 (http://wwwscirporg/ournal/sea) Quantization of Rough Set Based Reduction Bing Li *, Peng Tang, Tommy
More informationENTROPIES OF FUZZY INDISCERNIBILITY RELATION AND ITS OPERATIONS
International Journal of Uncertainty Fuzziness and Knowledge-Based Systems World Scientific ublishing Company ENTOIES OF FUZZY INDISCENIBILITY ELATION AND ITS OEATIONS QINGUA U and DAEN YU arbin Institute
More informationEasy Categorization of Attributes in Decision Tables Based on Basic Binary Discernibility Matrix
Easy Categorization of Attributes in Decision Tables Based on Basic Binary Discernibility Matrix Manuel S. Lazo-Cortés 1, José Francisco Martínez-Trinidad 1, Jesús Ariel Carrasco-Ochoa 1, and Guillermo
More informationOn rule acquisition in incomplete multi-scale decision tables
*Manuscript (including abstract) Click here to view linked References On rule acquisition in incomplete multi-scale decision tables Wei-Zhi Wu a,b,, Yuhua Qian c, Tong-Jun Li a,b, Shen-Ming Gu a,b a School
More informationHome Page. Title Page. Page 1 of 35. Go Back. Full Screen. Close. Quit
JJ II J I Page 1 of 35 General Attribute Reduction of Formal Contexts Tong-Jun Li Zhejiang Ocean University, China litj@zjou.edu.cn September, 2011,University of Milano-Bicocca Page 2 of 35 Objective of
More informationarxiv: v1 [cs.ai] 25 Sep 2012
Condition for neighborhoods in covering based rough sets to form a partition arxiv:1209.5480v1 [cs.ai] 25 Sep 2012 Abstract Hua Yao, William Zhu Lab of Granular Computing, Zhangzhou Normal University,
More informationRough Set Model Selection for Practical Decision Making
Rough Set Model Selection for Practical Decision Making Joseph P. Herbert JingTao Yao Department of Computer Science University of Regina Regina, Saskatchewan, Canada, S4S 0A2 {herbertj, jtyao}@cs.uregina.ca
More informationResearch Article Uncertainty Analysis of Knowledge Reductions in Rough Sets
e Scientific World Journal, Article ID 576409, 8 pages http://dx.doi.org/10.1155/2014/576409 Research Article Uncertainty Analysis of Knowledge Reductions in Rough Sets Ying Wang 1 and Nan Zhang 2 1 Department
More informationAn algorithm for induction of decision rules consistent with the dominance principle
An algorithm for induction of decision rules consistent with the dominance principle Salvatore Greco 1, Benedetto Matarazzo 1, Roman Slowinski 2, Jerzy Stefanowski 2 1 Faculty of Economics, University
More informationFeature Selection with Fuzzy Decision Reducts
Feature Selection with Fuzzy Decision Reducts Chris Cornelis 1, Germán Hurtado Martín 1,2, Richard Jensen 3, and Dominik Ślȩzak4 1 Dept. of Mathematics and Computer Science, Ghent University, Gent, Belgium
More informationResearch on Complete Algorithms for Minimal Attribute Reduction
Research on Complete Algorithms for Minimal Attribute Reduction Jie Zhou, Duoqian Miao, Qinrong Feng, and Lijun Sun Department of Computer Science and Technology, Tongji University Shanghai, P.R. China,
More informationRoman Słowiński. Rough or/and Fuzzy Handling of Uncertainty?
Roman Słowiński Rough or/and Fuzzy Handling of Uncertainty? 1 Rough sets and fuzzy sets (Dubois & Prade, 1991) Rough sets have often been compared to fuzzy sets, sometimes with a view to introduce them
More informationA new Approach to Drawing Conclusions from Data A Rough Set Perspective
Motto: Let the data speak for themselves R.A. Fisher A new Approach to Drawing Conclusions from Data A Rough et Perspective Zdzisław Pawlak Institute for Theoretical and Applied Informatics Polish Academy
More informationROUGH SETS THEORY AND DATA REDUCTION IN INFORMATION SYSTEMS AND DATA MINING
ROUGH SETS THEORY AND DATA REDUCTION IN INFORMATION SYSTEMS AND DATA MINING Mofreh Hogo, Miroslav Šnorek CTU in Prague, Departement Of Computer Sciences And Engineering Karlovo Náměstí 13, 121 35 Prague
More informationMultigranulation rough set: A multiset based strategy
Multigranulation rough set: A multiset based strategy Xibei Yang 1,2, Suping Xu 1, Huili Dou 1, Xiaoning Song 3, Hualong Yu 1, Jingyu Yang 4 1 School of Computer Science and Engineering, Jiangsu University
More informationTOPOLOGICAL ASPECTS OF YAO S ROUGH SET
Chapter 5 TOPOLOGICAL ASPECTS OF YAO S ROUGH SET In this chapter, we introduce the concept of transmissing right neighborhood via transmissing expression of a relation R on domain U, and then we study
More informationA Generalized Decision Logic in Interval-set-valued Information Tables
A Generalized Decision Logic in Interval-set-valued Information Tables Y.Y. Yao 1 and Qing Liu 2 1 Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 E-mail: yyao@cs.uregina.ca
More informationNaive Bayesian Rough Sets
Naive Bayesian Rough Sets Yiyu Yao and Bing Zhou Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 {yyao,zhou200b}@cs.uregina.ca Abstract. A naive Bayesian classifier
More informationREDUCTS AND ROUGH SET ANALYSIS
REDUCTS AND ROUGH SET ANALYSIS A THESIS SUBMITTED TO THE FACULTY OF GRADUATE STUDIES AND RESEARCH IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN COMPUTER SCIENCE UNIVERSITY
More informationTransversal and Function Matroidal Structures of Covering-Based Rough Sets
Transversal and Function Matroidal Structures of Covering-Based Rough Sets Shiping Wang 1, William Zhu 2,,andFanMin 2 1 School of Mathematical Sciences, University of Electronic Science and Technology
More informationApplication of Rough Set Theory in Performance Analysis
Australian Journal of Basic and Applied Sciences, 6(): 158-16, 1 SSN 1991-818 Application of Rough Set Theory in erformance Analysis 1 Mahnaz Mirbolouki, Mohammad Hassan Behzadi, 1 Leila Karamali 1 Department
More informationConcept Lattices in Rough Set Theory
Concept Lattices in Rough Set Theory Y.Y. Yao Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 E-mail: yyao@cs.uregina.ca URL: http://www.cs.uregina/ yyao Abstract
More informationi jand Y U. Let a relation R U U be an
Dependency Through xiomatic pproach On Rough Set Theory Nilaratna Kalia Deptt. Of Mathematics and Computer Science Upendra Nath College, Nalagaja PIN: 757073, Mayurbhanj, Orissa India bstract: The idea
More informationRough Sets and Conflict Analysis
Rough Sets and Conflict Analysis Zdzis law Pawlak and Andrzej Skowron 1 Institute of Mathematics, Warsaw University Banacha 2, 02-097 Warsaw, Poland skowron@mimuw.edu.pl Commemorating the life and work
More informationAndrzej Skowron, Zbigniew Suraj (Eds.) To the Memory of Professor Zdzisław Pawlak
Andrzej Skowron, Zbigniew Suraj (Eds.) ROUGH SETS AND INTELLIGENT SYSTEMS To the Memory of Professor Zdzisław Pawlak Vol. 1 SPIN Springer s internal project number, if known Springer Berlin Heidelberg
More informationRough Sets, Rough Relations and Rough Functions. Zdzislaw Pawlak. Warsaw University of Technology. ul. Nowowiejska 15/19, Warsaw, Poland.
Rough Sets, Rough Relations and Rough Functions Zdzislaw Pawlak Institute of Computer Science Warsaw University of Technology ul. Nowowiejska 15/19, 00 665 Warsaw, Poland and Institute of Theoretical and
More informationResearch Article The Uncertainty Measure of Hierarchical Quotient Space Structure
Mathematical Problems in Engineering Volume 2011, Article ID 513195, 16 pages doi:10.1155/2011/513195 Research Article The Uncertainty Measure of Hierarchical Quotient Space Structure Qinghua Zhang 1,
More informationARTICLE IN PRESS. Information Sciences xxx (2016) xxx xxx. Contents lists available at ScienceDirect. Information Sciences
Information Sciences xxx (2016) xxx xxx Contents lists available at ScienceDirect Information Sciences journal homepage: www.elsevier.com/locate/ins Three-way cognitive concept learning via multi-granularity
More informationNotes on Rough Set Approximations and Associated Measures
Notes on Rough Set Approximations and Associated Measures Yiyu Yao Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 E-mail: yyao@cs.uregina.ca URL: http://www.cs.uregina.ca/
More informationBanacha Warszawa Poland s:
Chapter 12 Rough Sets and Rough Logic: A KDD Perspective Zdzis law Pawlak 1, Lech Polkowski 2, and Andrzej Skowron 3 1 Institute of Theoretical and Applied Informatics Polish Academy of Sciences Ba ltycka
More informationDominance-based Rough Set Approach Data Analysis Framework. User s guide
Dominance-based Rough Set Approach Data Analysis Framework User s guide jmaf - Dominance-based Rough Set Data Analysis Framework http://www.cs.put.poznan.pl/jblaszczynski/site/jrs.html Jerzy Błaszczyński,
More informationFuzzy Integral Multiple Criteria Decision Making Method Based on Fuzzy Preference Relation on Alternatives
Journal of Systems Science and Information Jun., 2016, Vol. 4, No. 3, pp. 280 290 DOI: 10.21078/JSSI-2016-280-11 Fuzzy Integral Multiple Criteria Decision Making Method Based on Fuzzy Preference Relation
More informationRough operations on Boolean algebras
Rough operations on Boolean algebras Guilin Qi and Weiru Liu School of Computer Science, Queen s University Belfast Belfast, BT7 1NN, UK Abstract In this paper, we introduce two pairs of rough operations
More informationCharacterizing Pawlak s Approximation Operators
Characterizing Pawlak s Approximation Operators Victor W. Marek Department of Computer Science University of Kentucky Lexington, KY 40506-0046, USA To the memory of Zdzisław Pawlak, in recognition of his
More informationOn Rough Set Modelling for Data Mining
On Rough Set Modelling for Data Mining V S Jeyalakshmi, Centre for Information Technology and Engineering, M. S. University, Abhisekapatti. Email: vsjeyalakshmi@yahoo.com G Ariprasad 2, Fatima Michael
More informationSimilarity-based Classification with Dominance-based Decision Rules
Similarity-based Classification with Dominance-based Decision Rules Marcin Szeląg, Salvatore Greco 2,3, Roman Słowiński,4 Institute of Computing Science, Poznań University of Technology, 60-965 Poznań,
More informationROUGH NEUTROSOPHIC SETS. Said Broumi. Florentin Smarandache. Mamoni Dhar. 1. Introduction
italian journal of pure and applied mathematics n. 32 2014 (493 502) 493 ROUGH NEUTROSOPHIC SETS Said Broumi Faculty of Arts and Humanities Hay El Baraka Ben M sik Casablanca B.P. 7951 Hassan II University
More informationA Boolean Lattice Based Fuzzy Description Logic in Web Computing
A Boolean Lattice Based Fuzzy Description Logic in Web omputing hangli Zhang, Jian Wu, Zhengguo Hu Department of omputer Science and Engineering, Northwestern Polytechnical University, Xi an 7007, hina
More informationRough Set Approaches for Discovery of Rules and Attribute Dependencies
Rough Set Approaches for Discovery of Rules and Attribute Dependencies Wojciech Ziarko Department of Computer Science University of Regina Regina, SK, S4S 0A2 Canada Abstract The article presents an elementary
More informationFusing monotonic decision trees
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, FINAL VERSION 1 Fusing monotonic decision trees Yuhua Qian, Member, IEEE, Hang Xu, Jiye Liang, Bing Liu, Fellow, IEEE and Jieting Wang Abstract Ordinal
More informationClassification Based on Logical Concept Analysis
Classification Based on Logical Concept Analysis Yan Zhao and Yiyu Yao Department of Computer Science, University of Regina, Regina, Saskatchewan, Canada S4S 0A2 E-mail: {yanzhao, yyao}@cs.uregina.ca Abstract.
More informationA Swarm-based Rough Set Approach for Group Decision Support Systems
2009 Ninth International Conference on Hybrid Intelligent Systems A Swarm-based Rough Set Approach for Group Decision Support Systems Mingyan Zhao 1, Hongbo Liu 2,3, Ajith Abraham 2,3 and Emilio Corchado
More informationMathematical Approach to Vagueness
International Mathematical Forum, 2, 2007, no. 33, 1617-1623 Mathematical Approach to Vagueness Angel Garrido Departamento de Matematicas Fundamentales Facultad de Ciencias de la UNED Senda del Rey, 9,
More informationAdditive Preference Model with Piecewise Linear Components Resulting from Dominance-based Rough Set Approximations
Additive Preference Model with Piecewise Linear Components Resulting from Dominance-based Rough Set Approximations Krzysztof Dembczyński, Wojciech Kotłowski, and Roman Słowiński,2 Institute of Computing
More informationResearch Article Special Approach to Near Set Theory
Mathematical Problems in Engineering Volume 2011, Article ID 168501, 10 pages doi:10.1155/2011/168501 Research Article Special Approach to Near Set Theory M. E. Abd El-Monsef, 1 H. M. Abu-Donia, 2 and
More informationRough Set Approach for Generation of Classification Rules for Jaundice
Rough Set Approach for Generation of Classification Rules for Jaundice Sujogya Mishra 1, Shakti Prasad Mohanty 2, Sateesh Kumar Pradhan 3 1 Research scholar, Utkal University Bhubaneswar-751004, India
More informationResearch Article A Variable Precision Attribute Reduction Approach in Multilabel Decision Tables
e Scientific World Journal, Article ID 359626, 7 pages http://dx.doi.org/10.1155/2014/359626 Research Article A Variable Precision Attribute Reduction Approach in Multilabel Decision Tables Hua Li, 1,2
More informationCross-entropy measure on interval neutrosophic sets and its applications in Multicriteria decision making
Manuscript Click here to download Manuscript: Cross-entropy measure on interval neutrosophic sets and its application in MCDM.pdf 1 1 1 1 1 1 1 0 1 0 1 0 1 0 1 0 1 Cross-entropy measure on interval neutrosophic
More informationComputational Intelligence, Volume, Number, VAGUENES AND UNCERTAINTY: A ROUGH SET PERSPECTIVE. Zdzislaw Pawlak
Computational Intelligence, Volume, Number, VAGUENES AND UNCERTAINTY: A ROUGH SET PERSPECTIVE Zdzislaw Pawlak Institute of Computer Science, Warsaw Technical University, ul. Nowowiejska 15/19,00 665 Warsaw,
More informationAn Approach to Classification Based on Fuzzy Association Rules
An Approach to Classification Based on Fuzzy Association Rules Zuoliang Chen, Guoqing Chen School of Economics and Management, Tsinghua University, Beijing 100084, P. R. China Abstract Classification based
More informationSemantic Rendering of Data Tables: Multivalued Information Systems Revisited
Semantic Rendering of Data Tables: Multivalued Information Systems Revisited Marcin Wolski 1 and Anna Gomolińska 2 1 Maria Curie-Skłodowska University, Department of Logic and Cognitive Science, Pl. Marii
More informationA PRIMER ON ROUGH SETS:
A PRIMER ON ROUGH SETS: A NEW APPROACH TO DRAWING CONCLUSIONS FROM DATA Zdzisław Pawlak ABSTRACT Rough set theory is a new mathematical approach to vague and uncertain data analysis. This Article explains
More informationOutlier Detection Using Rough Set Theory
Outlier Detection Using Rough Set Theory Feng Jiang 1,2, Yuefei Sui 1, and Cungen Cao 1 1 Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences,
More informationStatistical Model for Rough Set Approach to Multicriteria Classification
Statistical Model for Rough Set Approach to Multicriteria Classification Krzysztof Dembczyński 1, Salvatore Greco 2, Wojciech Kotłowski 1 and Roman Słowiński 1,3 1 Institute of Computing Science, Poznań
More information1.4 Equivalence Relations and Partitions
24 CHAPTER 1. REVIEW 1.4 Equivalence Relations and Partitions 1.4.1 Equivalence Relations Definition 1.4.1 (Relation) A binary relation or a relation on a set S is a set R of ordered pairs. This is a very
More informationRough Sets for Uncertainty Reasoning
Rough Sets for Uncertainty Reasoning S.K.M. Wong 1 and C.J. Butz 2 1 Department of Computer Science, University of Regina, Regina, Canada, S4S 0A2, wong@cs.uregina.ca 2 School of Information Technology
More informationA Comparative Study of Different Order Relations of Intervals
A Comparative Study of Different Order Relations of Intervals Samiran Karmakar Department of Business Mathematics and Statistics, St. Xavier s College, Kolkata, India skmath.rnc@gmail.com A. K. Bhunia
More informationInternational Journal of Approximate Reasoning
International Journal of Approximate Reasoning 53 (22) 62 635 Contents lists available at SciVerse ScienceDirect International Journal of Approximate Reasoning journal homepage:www.elsevier.com/locate/ijar
More informationInternational Journal of Approximate Reasoning
International Journal of Approximate Reasoning 52 (2011) 231 239 Contents lists available at ScienceDirect International Journal of Approximate Reasoning journal homepage: www.elsevier.com/locate/ijar
More informationBulletin of the Transilvania University of Braşov Vol 10(59), No Series III: Mathematics, Informatics, Physics, 67-82
Bulletin of the Transilvania University of Braşov Vol 10(59), No. 1-2017 Series III: Mathematics, Informatics, Physics, 67-82 IDEALS OF A COMMUTATIVE ROUGH SEMIRING V. M. CHANDRASEKARAN 3, A. MANIMARAN
More informationResearch Article Structural Damage Identification Based on Rough Sets and Artificial Neural Network
e Scientific World Journal, Article ID 193284, 9 pages http://dx.doi.org/10.1155/2014/193284 Research Article Structural Damage Identification Based on Rough Sets and Artificial Neural Network Chengyin
More informationA Logical Formulation of the Granular Data Model
2008 IEEE International Conference on Data Mining Workshops A Logical Formulation of the Granular Data Model Tuan-Fang Fan Department of Computer Science and Information Engineering National Penghu University
More informationFoundations of Classification
Foundations of Classification J. T. Yao Y. Y. Yao and Y. Zhao Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 {jtyao, yyao, yanzhao}@cs.uregina.ca Summary. Classification
More informationOn Probability of Matching in Probability Based Rough Set Definitions
2013 IEEE International Conference on Systems, Man, and Cybernetics On Probability of Matching in Probability Based Rough Set Definitions Do Van Nguyen, Koichi Yamada, and Muneyuki Unehara Department of
More informationHigh Frequency Rough Set Model based on Database Systems
High Frequency Rough Set Model based on Database Systems Kartik Vaithyanathan kvaithya@gmail.com T.Y.Lin Department of Computer Science San Jose State University San Jose, CA 94403, USA tylin@cs.sjsu.edu
More informationParameters to find the cause of Global Terrorism using Rough Set Theory
Parameters to find the cause of Global Terrorism using Rough Set Theory Sujogya Mishra Research scholar Utkal University Bhubaneswar-751004, India Shakti Prasad Mohanty Department of Mathematics College
More informationClassification of Voice Signals through Mining Unique Episodes in Temporal Information Systems: A Rough Set Approach
Classification of Voice Signals through Mining Unique Episodes in Temporal Information Systems: A Rough Set Approach Krzysztof Pancerz, Wies law Paja, Mariusz Wrzesień, and Jan Warcho l 1 University of
More informationApplications of Some Topological Near Open Sets to Knowledge Discovery
IJACS International Journal of Advanced Computer Science Applications Vol 7 No 1 216 Applications of Some Topological Near Open Sets to Knowledge Discovery A S Salama Tanta University; Shaqra University
More informationNear approximations via general ordered topological spaces M.Abo-Elhamayel Mathematics Department, Faculty of Science Mansoura University
Near approximations via general ordered topological spaces MAbo-Elhamayel Mathematics Department, Faculty of Science Mansoura University Abstract ough set theory is a new mathematical approach to imperfect
More informationKnowledge Discovery. Zbigniew W. Ras. Polish Academy of Sciences, Dept. of Comp. Science, Warsaw, Poland
Handling Queries in Incomplete CKBS through Knowledge Discovery Zbigniew W. Ras University of orth Carolina, Dept. of Comp. Science, Charlotte,.C. 28223, USA Polish Academy of Sciences, Dept. of Comp.
More informationA NOVEL VIEW OF ROUGH SOFT SEMIGROUPS BASED ON FUZZY IDEALS. Qiumei Wang Jianming Zhan Introduction
italian journal of pure and applied mathematics n. 37 2017 (673 686) 673 A NOVEL VIEW OF ROUGH SOFT SEMIGROUPS BASED ON FUZZY IDEALS Qiumei Wang Jianming Zhan 1 Department of Mathematics Hubei University
More informationOn minimal models of the Region Connection Calculus
Fundamenta Informaticae 69 (2006) 1 20 1 IOS Press On minimal models of the Region Connection Calculus Lirong Xia State Key Laboratory of Intelligent Technology and Systems Department of Computer Science
More informationResearch Article Cost-Sensitive Attribute Reduction in Decision-Theoretic Rough Set Models
Mathematical Problems in Engineering, rticle ID 875918, 9 pages http://dx.doi.org/10.1155/2014/875918 Research rticle Cost-Sensitive ttribute Reduction in Decision-Theoretic Rough Set Models Shujiao Liao,
More informationA Rough Set Interpretation of User s Web Behavior: A Comparison with Information Theoretic Measure
A Rough et Interpretation of User s Web Behavior: A Comparison with Information Theoretic easure George V. eghabghab Roane tate Dept of Computer cience Technology Oak Ridge, TN, 37830 gmeghab@hotmail.com
More informationBhubaneswar , India 2 Department of Mathematics, College of Engineering and
www.ijcsi.org 136 ROUGH SET APPROACH TO GENERATE CLASSIFICATION RULES FOR DIABETES Sujogya Mishra 1, Shakti Prasad Mohanty 2, Sateesh Kumar Pradhan 3 1 Research scholar, Utkal University Bhubaneswar-751004,
More informationOn Some Structural Properties of Fuzzy Soft Topological Spaces
Intern J Fuzzy Mathematical Archive Vol 1, 2013, 1-15 ISSN: 2320 3242 (P), 2320 3250 (online) Published on 31 January 2013 wwwresearchmathsciorg International Journal of On Some Structural Properties of
More informationInternational Journal of Scientific & Engineering Research, Volume 6, Issue 3, March ISSN
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 969 Soft Generalized Separation Axioms in Soft Generalized Topological Spaces Jyothis Thomas and Sunil Jacob John
More informationUncertain Satisfiability and Uncertain Entailment
Uncertain Satisfiability and Uncertain Entailment Zhuo Wang, Xiang Li Department of Mathematical Sciences, Tsinghua University, Beijing, 100084, China zwang0518@sohu.com, xiang-li04@mail.tsinghua.edu.cn
More information960 JOURNAL OF COMPUTERS, VOL. 8, NO. 4, APRIL 2013
960 JORNAL OF COMPTERS, VOL 8, NO 4, APRIL 03 Study on Soft Groups Xia Yin School of Science, Jiangnan niversity, Wui, Jiangsu 4, PR China Email: yinia975@yahoocomcn Zuhua Liao School of Science, Jiangnan
More informationBelief Classification Approach based on Dynamic Core for Web Mining database
Third international workshop on Rough Set Theory RST 11 Milano, Italy September 14 16, 2010 Belief Classification Approach based on Dynamic Core for Web Mining database Salsabil Trabelsi Zied Elouedi Larodec,
More informationRough Soft Sets: A novel Approach
International Journal of Computational pplied Mathematics. ISSN 1819-4966 Volume 12, Number 2 (2017), pp. 537-543 Research India Publications http://www.ripublication.com Rough Soft Sets: novel pproach
More informationMatching Index of Uncertain Graph: Concept and Algorithm
Matching Index of Uncertain Graph: Concept and Algorithm Bo Zhang, Jin Peng 2, School of Mathematics and Statistics, Huazhong Normal University Hubei 430079, China 2 Institute of Uncertain Systems, Huanggang
More informationAN INTRODUCTION TO FUZZY SOFT TOPOLOGICAL SPACES
Hacettepe Journal of Mathematics and Statistics Volume 43 (2) (2014), 193 204 AN INTRODUCTION TO FUZZY SOFT TOPOLOGICAL SPACES Abdülkadir Aygünoǧlu Vildan Çetkin Halis Aygün Abstract The aim of this study
More informationThe size of decision table can be understood in terms of both cardinality of A, denoted by card (A), and the number of equivalence classes of IND (A),
Attribute Set Decomposition of Decision Tables Dominik Slezak Warsaw University Banacha 2, 02-097 Warsaw Phone: +48 (22) 658-34-49 Fax: +48 (22) 658-34-48 Email: slezak@alfa.mimuw.edu.pl ABSTRACT: Approach
More information1 Introduction Rough sets theory has been developed since Pawlak's seminal work [6] (see also [7]) as a tool enabling to classify objects which are on
On the extension of rough sets under incomplete information Jerzy Stefanowski 1 and Alexis Tsouki as 2 1 Institute of Computing Science, Poznań University oftechnology, 3A Piotrowo, 60-965 Poznań, Poland,
More informationFuzzy Modal Like Approximation Operations Based on Residuated Lattices
Fuzzy Modal Like Approximation Operations Based on Residuated Lattices Anna Maria Radzikowska Faculty of Mathematics and Information Science Warsaw University of Technology Plac Politechniki 1, 00 661
More informationA Chance-Constrained Programming Model for Inverse Spanning Tree Problem with Uncertain Edge Weights
A Chance-Constrained Programming Model for Inverse Spanning Tree Problem with Uncertain Edge Weights 1 Xiang Zhang, 2 Qina Wang, 3 Jian Zhou* 1, First Author School of Management, Shanghai University,
More informationNEAR GROUPS ON NEARNESS APPROXIMATION SPACES
Hacettepe Journal of Mathematics and Statistics Volume 414) 2012), 545 558 NEAR GROUPS ON NEARNESS APPROXIMATION SPACES Ebubekir İnan and Mehmet Ali Öztürk Received 26:08:2011 : Accepted 29:11:2011 Abstract
More information