基于粗糙集的决策树构造算法.pdfVIP

第36 卷 第11 期 计 算 机 工 程 2010 年6 月 Vol.36 No.11 Computer Engineering June 2010 ·软件技术与数据库· 文章编号：1000—3428(2010)11—0075—03 文献标识码：A 中图分类号：TP312 基于粗糙集的决策树构造算法 1 2 1 丁春荣 ，李龙澍 ，杨宝华 (1. 安徽农业大学信息与计算机学院，合肥 230036 ；2. 安徽大学计算机科学与技术学院，合肥 230039) 摘 要：针对ID3 算法构造决策树复杂、分类效率不高问题，基于粗糙集理论提出一种决策树构造算法。该算法采用加权分类粗糙度作为 节点选择属性的启发函数，与信息增益相比，能全面地刻画属性分类的综合贡献能力，并且计算简单。为消除噪声对选择属性和生成叶节 点的影响，利用变精度粗糙集模型对该算法进行优化。实验结果表明，该算法构造的决策树在规模与分类效率上均优于ID3 算法。 关键词：数据挖掘；粗糙集；可变精度粗糙集；决策树；加权分类粗糙度 Decision Tree Constructing Algorithm Based on Rough Set 1 2 1 DING Chun-rong , LI Long-shu , YANG Bao-hua (1. School of Information and Computer, Anhui Agricultural University, Hefei 230036; 2. School of Computer Science and Technology, Anhui University, Hefei 230039) 【Abstract 】Aiming at the problems of complex and low efficiency decision tree constructed by ID3, this paper proposes a decision tree classification algorithm based on rough set, which takes the weighted classification rough degree as the heuristic function of choosing attribute at a node. This heuristic function can synthetically measure contribution of an attribute for classification, and is simple in calculation. To eliminate the effect of noise data on choosing attributes and generating leaf nodes, a method using variable precision rough set model is used to optimize the algorithm