摘要
基于粗糙集的有关理论,提出了一种新的连续属性离散化方法·首先说明决策属性支持度的概念,再利用决策属性支持度作为反馈信息,提出一种领域独立的基于决策属性支持度的连续属性离散化算法·该算法能在保证决策表原始分类能力不变的前提下,提高约简效率·同时,各个属性拥有较少的分割区间,会使规则集合更加简洁·通过实例分析比较,说明该算法是非常有效的·
Based on theory of the rough set, a new method of discretization of continuous attributes was presented.The traditional rough set theory can only deal with the discrete attributes in database.The concept of decision attribute support degree was proposed. Using feedback information from decision attribute support degree a new method of discretization of continuous attributes was proposed based on independent domain. The suggested method can improve the efficiency of knowledge reduction when the original decision table keeps stable. At the same time, the rule sets are simple with less segmental interval. The approach is encouraging and effective.
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2003年第8期747-750,共4页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(70171056)
国家重点科技攻关项目(975620107)
关键词
粗糙集理论
决策表
决策属性支持度
离散化
数据挖掘
rough set theory
decision table
decision attribute support degree
discretization
data mining
