基于粗糙集理论的决策树分类方法  

A Classification Method Using Decision Trees Based on Rough Sets

作　　者：邹瑞芝[1] 罗可[1] 曾正良[1] 

机构地区：[1]长沙理工大学计算机与通信工程学院。湖南长沙410076

出　　处：《计算机工程与科学》2009年第10期112-114,共3页Computer Engineering & Science

基　　金：国家自然科学基金资助项目(10471036,60474070);湖南省科技计划项目基金资助项目(05FJ3074);湖南省教育厅重点项目(07A001)

摘　　要：决策树是数据挖掘中常用的分类方法。本文提出了基于粗糙集的决策树方法,利用粗糙集近似精确度来选择决策树的根节点,分支由分类产生。该方法计算简单,易于理解。本文还提出用悲观剪枝法简化决策树,提高决策树的预测与分类能力。实例说明了本文方法均简单有效。Decision tree is a usual method of classification in data mining. The decision tree method on the basis of RS is proposed in this paper, which chooses the root node on the basis of approximation quality, and produces branches via classification. The method is simple in calculation and easy-to.understand. In order to improve the efficiency of decision tree classification and forecast, the paper also proposes a pessimistic pruning algorithm for decision tree simplification. The example shows that the methods are simple and effective.

关 键 词：粗糙集 决策树 近似精确度 悲观剪枝 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]