期刊文献+

ID3算法的合理性证明及实验分析 被引量:1

The Proof of Rationality of ID3 Algorithm and Experimental Analysis
下载PDF
导出
摘要 研究一个属性的某几个属性值并的权熵之和与该属性单个属性值的权熵之和的关系,从理论上证明一个属性的某几个属性值并的权熵之和不小于该属性单个属性值的权熵之和.为ID3算法的合理性提供理论基础.实验结果证明结论正确. The relation between the weighted entropy of the union of several attribute values and the sum of the weighted entropy of the single attribute value is studied. The proof is given for the conclusion that the weighted entropy of the union of several attribute values is not less than the sum of the weighted entropy of the single attribute value, and the theoretical foundation for ID3 algorithm is presented. The results of the experiment conform the conclusion.
出处 《保定学院学报》 2008年第4期24-27,共4页 Journal of Baoding University
基金 河北农业大学非生命学科与新兴学科科研发展基金(FSY200739)
关键词 信息熵 ID3算法 决策树 条件属性 决策属性 entropy ID3 algorithm decision tree conditional attributes decision attributes
  • 相关文献

参考文献4

二级参考文献6

  • 1Han J, Kambr M. Data Mining: Concepts and Techniques. Morgan Kaufmann Publishers, 2001:279-333 被引量:1
  • 2Ruggieri S. Efficient C4.5. IEEE Transactions on Knowledge and Data Engineering, 2002, 14(2):438-444 被引量:1
  • 3Breiman L, Friedman JH, Olshen RA, et al. Classification and Regression Trees. Chapman & Hall(Wadsworth, Inc.): New York, 1984 被引量:1
  • 4Mehta M, Agrawal R, Rissancn J. SLIQ: A Fast Scalable Classifier for Data Mining. Research Report, IBM Almaden Research Center, San Jose, California, 1995 被引量:1
  • 5Shafer J, Agrawal R, Mehta M. SPRINT: A Scalable Parallel Classifier for Data Mining. Research Report, IBM Almaden Research Center,San Jose, California, 1996 被引量:1
  • 6Rastogi R, Shim K. PUBLIC: A Decision Tree Classifier that Integrates Building and Pruning. Technical Report, Bell Laboratories, Murray Hill, 1998 被引量:1

共引文献111

同被引文献3

引证文献1

二级引证文献9

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部