摘要
数据挖掘是一种重要的数据分析方法,决策树是数据挖掘中的一种主要技术,如何构造出最优决策树是许多研究者关心的问题。本文通过Rough集方法对决策表进行属性约简和属性值约简,去除决策表中与决策无关的冗余信息。在简化的决策表基础上构造近似最优决策树,本文给出了近似最优决策树的生成算法,并通过实例说明。
Data mining is an important data analysis method,and decision tree is one of main techniques in data mining.A problem concerned by many researches is how to construct a decision tree.In this paper,based on rough set,an attribute and attribute value reduction can be used to delete redundant decision information in a decision table.On the basis of compact information an approximate optimal decision tree is built.An Algorithm to generate decision tree is proposed in this paper,and it is illustrated by using a real example.
出处
《微计算机信息》
北大核心
2007年第01Z期296-297,282,共3页
Control & Automation
基金
江西省自然科学基金(JXPNSF-0311101
江西省教育厅科技计划项目(赣教技字[2006]31号)
