首页 | 官方网站   微博 | 高级检索  
     

基于属性间交互信息的模糊ID3算法的扩展
引用本文:王熙照,谢竞博.基于属性间交互信息的模糊ID3算法的扩展[J].复旦学报(自然科学版),2004,43(5):777-780.
作者姓名:王熙照  谢竞博
作者单位:河北大学,数学与计算机学院,保定071002
摘    要:模糊ID3算法是模糊决策树归纳中比较普遍和有效的启发式算法.以模糊ID3算法为例,分析了属性之间的冗余信息对构建模糊决策树的影响,并提出一个扩展算法,要求所选择的测试属性不仅和类的交互信息较大,而且和祖先节点上用过的属性之间的交互信息较小.实验结果表明:扩展算法优于模糊ID3算法

关 键 词:ID3算法  交互  互信息  属性  扩展  模糊决策树  启发式算法  构建  要求  实验结果

An Extended Fuzzy-ID3 Based on the Mutual Information between Attributes
Abstract.An Extended Fuzzy-ID3 Based on the Mutual Information between Attributes[J].Journal of Fudan University(Natural Science),2004,43(5):777-780.
Authors:Abstract
Abstract:Fuzzy-ID3 algorithm is a popular and efficient heuristic algorithm infuzzy decision tree induction. Taking Fuzzy-ID3 algorithm as the example, it analyzes the effect of the redundancy of attributes on the generation of fuzzy decision tree, and then proposes an extended version in which the testing attribute is selected based on not only the more mutual information between a candidate attribute and the class but also the less mutual information between a candidate attribute and the selected attributes on the same branch. The experimental result indicates that it has advantages over Fuzzy-ID3.
Keywords:machine learning  fuzzy decision tree  Fuzzy-ID3 algorithm  mutual information
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号