基于属性重要度的随机决策树学习算法  

A random decision tree algorithm based on attribute significance

作　　者：胡学钢[1] 李楠[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6期681-685,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(60573174);安徽省自然科学基金资助项目(050420207)

摘　　要：文章提出一种基于属性重要度的随机决策树构造算法ASRDT,该算法利用粗糙集理论计算每个属性的属性重要度,提升重要属性的影响因子,使得建树过程中随机选择属性时,不同属性之间的区分度得以体现,从而显著提高了算法的抗干扰能力,使ASRST在保持原有RDT算法优点的基础上,更具有良好的分类准确率及稳定性。An attribute significance-based random decision tree（ASRDT） algorithm is proposed in this paper, which appends significance for every attribute by the rough set theory to promote the impact factor of significant attributes. The appending of attribute significance distinguishes attributes in building,thus notably improving the ability of anti-jamming of the algorithm. The ASRDT algorithm not only has all the advantages of the random decision tree（RDT） algorithm but also possesses evidently higher correctness and better stability of classification than the RDT algorithm.

关 键 词：数据挖掘 分类 随机决策树 属性重要度 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程]