摘要
决策树算法是一种重要的数据挖掘方法,ID3算法是最具影响的一种决策树生成算法。介绍了粗集理论的相关概念和传统的ID3算法基本原理,提出了一种以粗集论中的属性依赖度为基础的ID3改进算法,克服了传统ID3算法对取值较多属性的依赖,并通过实例验证该算法的高效性和精确性。改进算法对不同领域中分类预测方向上的数据挖掘均具有一定的参考价值。
The algorithm of decision tree is an important method of data mining,and ID3 is one of the most influential decision tree generation algorithm.The related concepts of rough sets theory and the basic principles of traditional ID3 algorithm are introduced.An improved ID3 algorithm based on attribute dependence of rough sets theory is proposed,which overcomes traditional ID3 algorithm of depending on attribute of more values.The efficiency and accuracy of the algorithm are verified through an example.The improved algorithm has some reference value for classification and prediction of data mining in different fields.
出处
《河南科技大学学报（自然科学版）》
CAS
北大核心
2010年第1期42-45,共4页
Journal of Henan University of Science And Technology:Natural Science
基金
河南省自然科学基金项目(072300410180)
河南省高校科技创新人才支持计划项目(2008HASTIT029)
河南省教育厅科技攻关项目(2007520033)
关键词
粗集
ID3算法
属性依赖度
数据挖掘
Rough sets
ID3 algorithm
Attribute dependence
Data mining
