基于粗集的混合变量决策树构造算法研究  

On the decision tree inductive algorithm based on the rough set theory

作　　者：胡学钢[1] 张冬艳[1] 

机构地区：[1]合肥工业大学计算机与信息学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2007年第3期257-260,共4页Journal of Hefei University of Technology：Natural Science

基　　金：安徽省自然科学基金资助项目(050420207)

摘　　要：文章提出混合变量决策树结构,并在此基础上提出基于粗集理论的混合变量决策树构造算法RSH2,算法在每个结点选择尽可能少的属性明确划分尽可能多的实例,减小了决策树规模,且易于理解。将RSH2算法与ID3算法及基于粗集的单变量决策树算法HACRs进行实验比较,结果表明该算法有良好性能。The structure of the hybrid decision tree and the constructing algorithm for the structure are proposed in this paper. The hybrid decision tree algorithm RSH2, which is based on the rough set theory, selects the attributes as few as possible which can classify the instants as many as possible. Thus, the scale of the decision tree will be diminished and the tree will be easier to understand. In addition, the conditional information entropy based reduction algorithm is used before constructing the tree so that time will be saved. The comparison among the algorithms RSH2, ID3 and HACRs based on the rough set is made with experiments, and the RSH2 algorithm is proved to have good performance.

关 键 词：单变量决策树 多变量决策树 粗糙集合 归纳学习 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]