南京大学学报(自然科学版) ›› 2022, Vol. 58 ›› Issue (2): 255–263.doi: 10.13232/j.cnki.jnju.2022.02.009

• • 上一篇    

含属性加权模糊序决策信息系统的近似约简

徐伟华(), 孔子默, 陈曜琦   

  1. 西南大学人工智能学院,重庆,400715
  • 收稿日期:2021-10-22 出版日期:2022-03-30 发布日期:2022-04-02
  • 通讯作者: 徐伟华 E-mail:chxuwh@gmail.com
  • 作者简介:E⁃mail:chxuwh@gmail.com
  • 基金资助:
    国家自然科学基金(61976245)

Approximate reduction of fuzzy ordered decision information system with attribute weighting

Weihua Xu(), Zimo Kong, Yaoqi Chen   

  1. College of Artificial Intelligent,Southwest University,Chongqing,400715,China
  • Received:2021-10-22 Online:2022-03-30 Published:2022-04-02
  • Contact: Weihua Xu E-mail:chxuwh@gmail.com

摘要:

为保证关键属性在属性约简时能够被保留,可对信息系统的属性进行加权,从而提高关键属性的影响力.基于此,在属性加权的模糊序决策信息系统中建立了上、下近似约简的模型,得到两种约简的判定定理,并且给出求解上、下近似约简的辨识矩阵以及约简方法.最后,通过实例验证了该约简方法的有效性.

关键词: 辨识矩阵, 模糊序决策信息系统, 属性加权, 近似约简

Abstract:

In order to ensure that the key attributes can be retained during attribute reduction process,attributes of the information system can be weighted,so as to improve the influence of the key attributes. Based on this,this paper establishes the models of upper and lower approximate reduction in fuzzy order decision information system with attribute weighting,obtains the judgment theorems of two kinds of reduction,and gives the identification matrix and reduction method for solving the upper and lower approximate reduction. Finally,an example is given to verify the effectiveness of the reduction method.

Key words: identification matrix, fuzzy ordered decision information system, attribute weighting, approximate reduction

中图分类号: 

  • TP18

表1

一个模糊序决策信息系统"

Ua1(ω1=0.1)a2(ω2=0.4)a3(ω3=0.5)d
x10.70.30.93
x20.50.10.11
x30.50.302
x40.50.20.45
x50.70.40.54

表2

表1的上近似辨识矩阵"

x1x2x3x4x5
x1?????
x2??a3??
x3?????
x4?????
x5?????

表3

表1的下近似辨识矩阵"

x1x2x3x4x5
x1?a3a3??
x2?????
x3?a2???
x4?a3a3??
x5?a3a3??

表4

实验数据集总览"

No.数据集样本数特征数分类数
1

Caesarian Section

Classification Dataset

7942
2iris15043
3wine178133
4Connectionist Bench208602
5seed21073
6

Blood Transfusion

Service Center

74842
7audit_risk776262
8banknote authentication137232

表5

在SVM与KNN下所得上近似约简的分类精度"

数据集SVMKNN
平均84.27%±2.91%81.08%±3.06%

Caesarian Section

Classification Dataset

61.90%±5.83%68.75%±4.93%
iris99.17%±0.55%96.67%±0.72%
wine95.71%±0.63%97.14%±0.47%
Connectionist Bench84.85%±3.26%76.16%±2.96%
seed71.86%±4.38%61.90%±6.47%

Blood Transfusion

Service Center

76.76%±4.72%75.33%±4.37%
audit_risk86.13%±2.36%81.41%±2.74%
banknote authentication97.81%±1.51%91.27%±1.79%

表6

在SVM与KNN下所得下近似约简的分类精度"

数据集SVMKNN
平均87.51%±2.97%73.01%±4.45%

Caesarian Section

Classification Dataset

74.60%±5.47%43.75%±7.46%
iris74.17%±4.68%63.33%±4.37%
wine91.43%±3.62%65.71%±4.73%
Connectionist Bench88.79%±4.38%69.52%±6.24%
seed99.04%±1.49%85.71%±2.84%

Blood Transfusion

Service Center

78.92%±5.37%77.33%±5.22%
audit_risk95.81%±1.74%87.82%±2.68%
banknote authentication97.35%±1.37%90.91%±2.03%
1 Taj S M, Sudha M, Kumaravel A. Predicting heart failure using data mining with rough set theory and Fuzzy Petri Net. Journal of Physics:Conference Series,2021(1724):012033.
2 Wang G Q, Li T R, Zhang P F,et al. Double?local rough sets for efficient data mining. Information Sciences,2021(571):475-498.
3 Wang Y F. Mining stock price using fuzzy rough set system. Expert Systems with Applications,2003,24(1):13-23.
4 Ji W T, Pang Y, Jia X Y,et al. Fuzzy rough sets and fuzzy rough neural networks for feature selection:A review. WIREs Data Mining and Knowledge Discovery,2021,11(3):e1402.
5 高建来,运士伟,张永胜. 融合粗糙集与球形支持向量机的多分类识别. 河南科技大学学报:自然科学版,2011,32(5):77-80.
Gao J L, Yun S W, Zhang Y S. Multi?classification pattern recognition based on fusion of rough set and sphere?structured support vector machines. Journal of Henan University of Science and Technology:Natural Science,2011,32(5):77-80.
6 王江荣,黄建华,罗资琴,等. 基于粗糙集的Logistic回归模型在矿井突水模式识别中的应用. 煤田地质与勘探,2015,43(6):70-74.
Wang J R, Huang J H, Luo Z Q,et al. Application of logistic regression model based on rough set in recognition of mine water inrush pattern. Coal Geology & Exploration,2015,43(6):70-74.
7 Csajbók Z. Partial approximative set theory:A generalization of the rough set theory∥2010 International Conference of Soft Computing and Pattern Recognition. Cergy?Pontoise,France:IEEE,2010:51-56.
8 Zhou H, Wang Z, Shi C Q,et al. Study of fault diagnosis distribution network based on rough set and artificial intelligence. Journal of Physics:Conference Series,2021(1754):012204.
9 Trabelsi S, Elouedi Z, Lingras P. Belief rough set classifier∥Proceedings of the 22nd Canadian Conference on Artificial Intelligence:Advances in Artificial Intelligence. Springer Berlin Heidelberg,2009:257-261.
10 徐伟华. 序信息系统与粗糙集. 北京:科学出版社,2013.
11 袁修久,张文修. 模糊目标信息系统的属性约简. 系统工程理论与实践,2004,24(5):116-120,125.
Yuan X J, Zhang W X. Attribute reductions in fuzzy inconsistent information systems. Systems Engineering?Theory & Practice,2004,24(5):116-120,125.
12 Jensen R, Shen Q. Fuzzy–rough attribute reduction with application to web categorization. Fuzzy Sets and Systems,2004,141(3):469-485.
13 Jensen R, Shen Q. Fuzzy?rough data reduction with ant colony optimization. Fuzzy Sets and Systems,2005,149(1):5-20.
14 Pawlak Z. Rough Sets:Theoretical aspects of reasoning about data. Norwell:Kluwer Academic Publishers,1992.
15 Pawlak Z, Grzymala?Busse J, Slowinski R,et al. Rough sets. Communications of the ACM,1995,38(11):88-95.
16 张文修,米据生,吴伟志. 不协调目标信息系统的知识约简. 计算机学报,2003,26 (1):12-18.
Zhang W X, Mi J S, Wu W Z. Knowledge reductions in inconsistent information systems. Chinese Journal of Computers,2003,26(1):12-18.
17 王珏,苗夺谦,周育健. 关于Rough Set理论与应用的综述. 模式识别与人工智能,1996,9(4):337-344.
Wang J, Miao D Q, Zhou Y J. Rough set theory and its application:A Survey. PR & AI,1996,9(4):337-344.
[1] 敬思惠,秦克云*. 决策系统基于特定决策类的上近似约简[J]. 南京大学学报(自然科学版), 2018, 54(4): 804-.
[2] 陶玉枝1,2,赵仕梅1,2,谭安辉1,2*. 一种基于决策表约简的集覆盖问题的近似解法[J]. 南京大学学报(自然科学版), 2018, 54(4): 821-.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!