摘要
针对基于粗糙集理论的数据处理中存在冗余信息的问题,提出了冗余规则处理架构。利用粗糙集理论中属性约简的概念,把规则库中的单个规则作为条件属性建立新的决策表,通过规则约简剔除冗余规则。基于粗糙集理论中属性核的作用,把规则在多个约简中的出现频度作为规则重要性度量标准。实验结果表明,规则处理方法能够在有效剔除冗余规则的基础上,正确地对剩余规则进行重要性排序,对决策制定提供可靠的依据。
As redundant information existed in the rough set theory based data process, a method which processes redundant rules is presented. Firstly, rules as conditional attribute to construct a new decision table, and the concept of attribute reduct in rough set theory is utilized to reject redundant rules; Seconely, the frequency of each rule in the reduct is considered as the rule importance measure based on the attribute core in the rough set theory. Experimental results show that rule process method can reject redundant rules efficiently, measure the importance of rule correctly and provide reliable evidence for right decision_
出处
《计算机工程与设计》
CSCD
北大核心
2014年第1期21-25,81,共6页
Computer Engineering and Design
基金
国家973重点基础研究发展计划基金项目(2011CB311801)
国家863高技术研究发展技术基金项目(2012AA012704)
关键词
粗糙集
属性约简
属性核
冗余规则
数据处理
rough set
attribute reduct
attribute core
redundant rules
data process
