| 基于用户兴趣度和MID3决策树改进方法 |
| |
| 引用本文： | 王永梅,胡学钢.基于用户兴趣度和MID3决策树改进方法[J].计算机工程与应用,2011,47(27):155-157. |
| |
| 作者姓名： | 王永梅 胡学钢 |
| |
| 作者单位： | 1.安徽农业大学 信息与计算机学院，合肥 2300362.合肥工业大学 计算机与信息学院，合肥 230009 |
| |
| 基金项目： | 国家自然科学基金(No.30971691)~~ |
| |
| 摘 要： | 决策树是数据挖掘的重要方法，通常用来形成分类器和预测模型。分析ID3算法和现有的ID3改进算法所存在的问题。提出一种合理且可靠的MID3的改进算法，即针对MID3算法倾向于取值较多属性的缺点，引入了用户兴趣度对算法进行改进，改进后的MID3算法既可以在一定程度上解决多值偏向问题，也可以考虑决策树的两层节点。并通过实验与其他ID3改进算法进行比较，实验结果表明，改进后的MID3算法是有效的。
|
| 关 键 词： | 决策树 MID3算法 用户兴趣度 |
| 修稿时间： | |
Improved decision tree algorithm based on user-interest and MID3 |
| |
| WANG Yongmei,HU Xuegang.Improved decision tree algorithm based on user-interest and MID3[J].Computer Engineering and Applications,2011,47(27):155-157. |
| |
| Authors: | WANG Yongmei HU Xuegang |
| |
| Affiliation: | 1.School of Information and Computer，Anhui Agricultural University，Hefei 230036，China2.Institute of Computer & Information，Hefei University of Technology，Hefei 230009，China |
| |
| Abstract: | Decision tree is an important approach in data mining,which can be used to form classification and predictive model.In the present essay,ID3 algorithm and its improved algorithms at present are analyzed,and then a more reliable and reasonable approach is introduced in which user-interest is involved to solve the problem of multi-value and the double nodes as well.The improved MID 3 algorithm is proved efficient by the experiment in which several ID3 algorithms are compared. |
| |
| Keywords: | decision tree MID3 algorithm user-interest
|
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程与应用》浏览原始摘要信息 |
|
点击此处可从《计算机工程与应用》下载全文 |
