属性值和属性变化的增量属性约简算法

doi:10.6040/j.issn.1671-9352.1.2019.006

摘要/Abstract

摘要： 针对决策信息系统属性增加且属性值发生细化的情况下如何快速更新属性约简的问题,探讨了基于矩阵方法计算决策信息系统相对知识粒度的增量更新机理,设计了属性增加且属性值发生细化的矩阵增量约简算法。当决策表中的属性值细化且决策表中属性增加时,所提出的增量约简算法与非增量约简算法及其他增量约简算法相比,约简的分类精度变化不大,但是能够大大缩短计算约简的运行时间。最后利用一些UCI数据集做了大量仿真实验,仿真结果验证了所给出的动态属性约简算法能够有效地解决动态数据约简的问题。

关键词: 粗糙集, 增量学习, 属性约简, 知识粒度

Abstract: The incremental mechanisms to calculate relative knowledge granularity based on matrices are introduced when multiple attributes are added into decision system and attribute values refining. Then, the corresponding incremental attribute reduction method based on matrix is developed. Compared with the non-incremental attribute reduction algorithm and other incremental algorithms, the proposed incremental attribute reduction algorithm based on matrix can obtain a new reduct in a much shorter time. Finally, experiments on some data sets downloaded from UCI show that the proposed incremental attribute reduction method based on matrix is effective and efficient.

Key words: rough set, incremental learning, attribute reduction, knowledge granularity

中图分类号:

TP18

景运革,景罗希,王宝丽,程妮. 属性值和属性变化的增量属性约简算法[J]. 《山东大学学报(理学版)》, 2020, 55(1): 62-68.

JING Yun-ge, JING Luo-xi, WANG Bao-li, CHENG Ni. An incremental attribute reduction approach when attribute values and attributes of the decision system change dynamically[J]. JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE), 2020, 55(1): 62-68.

参考文献

[1] 苖夺谦, 范世栋. 知识粒度的计算及其应用[J].系统工程理论与实践, 2002, 22(1):48-56. MIAO Duoqian, FAN Shidong. The calculation of knowledge granulation and its application[J]. Systems Engineer-Theory & Practice, 2002, 22(1):48-56.
[2] 王国胤, 于洪, 杨大春. 基于条件熵的决策表约简[J].软件学报, 2002, 25(7):760-765. WANG Guoyin, YU Hong, YANG Dachun. Decision table reduction based on conditional information entropy[J]. Chinese Journal of Computer, 2002, 25(7):760-765.
[3] 刘少辉, 盛秋戬, 史忠植.一种新的快速计算正区域的方法[J].计算机研究与发展, 2003, 40(5):637-642. LIU Shaohui, SHENG Qiujian, SHI Zhongzhi. A new method for fast computing positive region[J]. Journal of Computer Research and Development, 2003, 40(5):637-642.
[4] ZENG A P, LI T R, LIU D, et al. A fuzzy rough set approach for incremental feature selection on hybrid information systems[J]. Fuzzy Sets and Systems, 2015, 258(1):39-60.
[5] SHU W H, SHEN H. Updating attribute reduct in incomplete decision systems with the variation of attribute set[J]. International Journal of Approximate Reasoning, 2014, 55(3):867-884.
[6] JING Y G, LI T R, HUANG J F, et al. An incremental attribute reduction approach based on knowledge granularity under the attribute generalization [J]. International Journal of Approximate Reasoning, 2016, 76:80-95.
[7] WANG F, LIANG J Y, QIAN Y H. Attribute reduction for dynamic data sets[J]. Applied Soft Computing, 2013, 13(1):676-689.
[8] CHEN D G, YANG Y Y, DONG Z. An incremental algorithm for attribute reduction with variable precision rough sets[J]. Applied Soft Computing, 2016, 45:129-149.
[9] 刘洋, 冯博琴, 周江卫.基于差别矩阵的增量式属性约简完备算法[J].西安交通大学学报, 2007,41(2):158-161. LIU Yang, FENG Boqin, ZHOU Jiangwei. Complete algorithm of increment for attribute reduction based on discernibility matrix[J]. Journal of Xian Jiaotong University, 2007, 41(2):158-161.
[10] 唐定勇, 景运革. 一种决策表属性值细化的正域约简算法[J].微电子学与计算机, 2015, 32(3):23-27. TANG Dingyong, JING Yunge. A reduction algorithm of positive domain for decision table based on values refining[J]. Microelectronics & Computer, 2015, 32(3):23-27.
[11] CHEN H M, LI T R, LUO C, et al. A rough set-based method for updating decision rules on attribute values coarsening and refining[J]. IEEE Transactions on Knowledge and Data Engineering, 2014, 26(12):2886-2899.
[12] 刘清. Rough set及Rough推理[M]. 北京:科学出版社, 2001. LIU Qing. Rough set and Rough reasoning[M]. Beijing: Science Press, 2001.
[13] 王磊, 叶军. 知识粒度计算的矩阵方法及其在属性约简中的应用[J]. 计算机工程与科学, 2013, 35(3):98-102. WANG Lei, YE Jun. Matrix-based approach for calculating knowledge granulation and its application in attribute reduction[J]. Computer Engineering & Science, 2013, 35(3):98-102.
[14] 王磊, 洪志全, 万旎. 属性值变化时变精度粗糙集模型中近似集动态更新的矩阵方法研究[J]. 计算机应用研究, 2013, 30(7):2011-2013. WANG Lei, HONG Zhiquan, WAN Ni. Research on matrix-based incremental method for updating approximations under variation of attribute values in variable precision rough set[J]. Application Research of Computers, 2013, 30(7):2011-2013.
[15] 郑诚, 王波, 洪彤彤. 关系矩阵的知识粒度增量式属性约简[J]. 小型微型计算机系统, 2018, 39(5):1000-1004. ZHENG Cheng, WANG Bo, HONG Tongtong. Incremental attribute reduction based on relational matrix[J]. Journal of Chinese Computer Systems, 2018, 39(5):1000-1004.

相关文章 15

[1]	张海洋,马周明,于佩秋,林梦雷,李进金. 多粒度粗糙集近似集的增量方法[J]. 《山东大学学报(理学版)》, 2020, 55(1): 51-61.
[2]	万青,马盈仓,魏玲. 基于多粒度的多源数据知识获取[J]. 《山东大学学报(理学版)》, 2020, 55(1): 41-50.
[3]	曾雪强,叶震麟,左家莉,万中英,吴水秀. 一种块增量偏最小二乘算法[J]. 《山东大学学报(理学版)》, 2019, 54(3): 93-101.
[4]	郑荔平,胡敏杰,杨红和,林耀进. 基于粗糙集的协同过滤算法研究[J]. 《山东大学学报(理学版)》, 2019, 54(2): 41-50.
[5]	李金海,吴伟志,邓硕. 形式概念分析的多粒度标记理论[J]. 《山东大学学报(理学版)》, 2019, 54(2): 30-40.
[6]	张恩胜. 区间集概念格属性约简的组成与结构[J]. 山东大学学报（理学版）, 2018, 53(8): 17-24.
[7]	左芝翠,张贤勇,莫智文,冯林. 基于决策分类的分块差别矩阵及其求核算法[J]. 山东大学学报（理学版）, 2018, 53(8): 25-33.
[8]	李同军,黄家文,吴伟志. 基于相似关系的不完备形式背景属性约简[J]. 山东大学学报（理学版）, 2018, 53(8): 9-16.
[9]	张晓,杨燕燕. 覆盖决策系统的规则提取和置信度保持的属性约简算法[J]. 《山东大学学报(理学版)》, 2018, 53(12): 120-126.
[10]	胡谦,米据生,李磊军. 多粒度模糊粗糙近似算子的信任结构与属性约简[J]. 山东大学学报（理学版）, 2017, 52(7): 30-36.
[11]	李丽,管涛,林和. 基于泛系算子的泛系混合并联粗糙集模型[J]. 山东大学学报（理学版）, 2017, 52(7): 22-29.
[12]	汪小燕,沈家兰,申元霞. 基于加权粒度和优势关系的程度多粒度粗糙集[J]. 山东大学学报（理学版）, 2017, 52(3): 97-104.
[13]	陈雪,魏玲,钱婷. 基于AE-概念格的决策形式背景属性约简[J]. 山东大学学报（理学版）, 2017, 52(12): 95-103.
[14]	黄伟婷,赵红,祝峰. 代价敏感属性约简的自适应分治算法[J]. 山东大学学报（理学版）, 2016, 51(8): 98-104.
[15]	安秋生, 孔祥玉. 函数依赖与多值依赖的再研究[J]. 山东大学学报（理学版）, 2014, 49(08): 1-5.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed