摘要
粗糙集理论是一个新的数据挖掘方法，正越来越被人们所重视。其主要思想是保持分类能力不变的情况下，利用等价类，通过属性约简和决策规则约简，达到发掘知识并简化知识的目的。但是属性约简是一个NP问题，对属性的约简和决策规则的约简只能通过启发式算法实现。该文针对属性约简和决策规则约简，各提出了一个启发式算法。
Rough set is a new method of data mining. It is being recognized gradually. Its basic theory is utilizing equivalence relation class, through attribution reduction and decision rule reduction, obtaining knowledge and the reduction of knowledge with the same ability of classification. But the attribute reduction is a NP problem, the attribution reduction and decision rule reduction will be solved by method of elicitation. In this paper, two methods of elicitation for attribution reduction and decision rule reduction are proposed.
出处
《计算机工程》
CAS
CSCD
北大核心
2003年第12期5-7,共3页
Computer Engineering
基金
国家"863"计划项目(2001AA412260)
关键词
粗糙集
属性约简
决策规则约简
启发式
Rough set
Attribution reduction
Decision rule reduction
Method of elicitation
