基于可辨识集消解策略的覆盖决策系统约简算法  

Reduction Algorithm of Covering Decision Systems Based on Discernibility Set Dissolving

作　　者：林荣德[1] 李进金 陈东晓 黄建新[1] 施晗娟[1] LIN Rongde;LI Jinjin;CHEN Dongxiao;HUANG Jianxin;SHI Hanjuan(Fujian Province University Key Laboratory of Computational Science,School of Mathematical Science,Huaqiao University,Quanzhou 362000,China;School of Mathematics and Statistics,Minnan Normal University,Zhangzhou 363000,China)

机构地区：[1]华侨大学计算科学福建省高校重点实验室数学科学学院,福建泉州362000 [2]闽南师范大学数学与统计学院,福建漳州363000

出　　处：《郑州大学学报（理学版）》2020年第1期8-15,共8页Journal of Zhengzhou University:Natural Science Edition

基　　金：国家自然科学基金项目(11871259);福建省自然科学基金项目(2019J01748,2019J01751,2016J01304);福建省高校创新团队发展计划;泉州市高层次人才团队项目(2017ZT012)

摘　　要：通过分析覆盖决策系统中的协调覆盖集与对象的可辨识集之间的关系,提出覆盖相对于可辨识集簇的辨识重要度和辨识影响度所构成的覆盖辨识能力,以及覆盖对可辨识集簇进行消解操作的概念。给出一种算法从可辨识集簇中提取最大辨识能力的覆盖,并用该覆盖对可辨识集簇进行逐次消解迭代计算,高效地将可辨识集簇完全消解来得到约简覆盖集。最后给出的UCI数据集上的约简实验结果表明了算法的有效性。Based on the close relationship between consistent cover sets and discernibility sets of objects in covering decision systems,the definition of discerning capacity of a cover,which consisted of discerning significance and discerning scalability relative to the family of discernibility set in a covering decision system was given.And then the dissolving operation on the family of discernibility set with a given cover was defined,and thus the reduction algorithm was proposed,which repeatedly selected a cover with maximum discerning capacity to be applied in the dissolving operation on the family of discernibility sets,until the family of discernibility sets became empty and the reduced cover set of the covering decision system came out.In order to verify practical validity of the algorithm,some UCI datasets were applied in the experiments,and they demonstrated the algorithm works well.

关 键 词：可辨识集消解 辨识重要度 辨识影响度 覆盖决策系统 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]