基于决策等价性的决策表属性集分解研究  

Attribute Set Decomposition of Decision Table Based on Decision Making Equivalence

作　　者：张政[1] 胡沛[1] 

机构地区：[1]南阳理工学院软件学院,南阳473000

出　　处：《计算机与数字工程》2016年第11期2224-2228,共5页Computer & Digital Engineering

摘　　要：决策表属性集分解是处理决策大型决策表数据复杂性,提高数据分析的一种有效手段,已得到深入研究。但在属性集分解过程中,有可能出现决策规则的泛化,从而导致从原决策表与从子决策表得到的规则不一致性。论文深入研究了决策表属性集分解的等价性问题,从保持决策表等价性和提高子表分类质量的角度,提出了基于决策等价的决策表属性集分解方法,并与现有的属性集分解方法做了比较。Attribute set decomposition of decision table as a useful method deals with data complexities of the large decision tables and improves data analysis.It had in-depth studied by many scholars.But it may appears decision rules generalization during the attribute set decomposition,so it may led differences of the rules from the original decision table and the child decision tables.The paper went deep into the equivalence of attribute set decomposition,and proposed novel attribute set decomposition approaches from keeping the equivalence of decision table and improving the classification rate in the child decision tables and compared the methods with existing methods of attribute set decomposition.

关 键 词：决策表 属性集 分解 等价性 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]