| 选定根节点的TAN分类器 |
| |
| 引用本文： | 王建林,王学玲.选定根节点的TAN分类器[J].滨州学院学报,2008,24(3). |
| |
| 作者姓名： | 王建林 王学玲 |
| |
| 作者单位： | 滨州学院,计算机科学技术系,山东,滨州,256603 |
| |
| 摘 要： | 通过分析树扩张型朴素贝叶斯(TAN)分类器,提出了一种选定根节点方法,此方法基于概率论的观点,构造生成树时选择两依赖属性中个数较大的属性作为根节点.依此对TAN分类器的构造方法进行了改进(改进的分类器简称CRTAN),将该分类方法与朴素贝叶斯分类器和TAN分类器进行了实验比较.实验结果表明,此分类方法的分类性能要优于TAN分类器.
|
| 关 键 词： | 分类器 树扩张型朴素贝叶斯 依赖关系 |
Choose Root Tree Algorithm Based TAN Classifiers |
| |
| WANG Jian-lin,WANG Xue-ling.Choose Root Tree Algorithm Based TAN Classifiers[J].Journal of Binzhou University,2008,24(3). |
| |
| Authors: | WANG Jian-lin WANG Xue-ling |
| |
| Abstract: | In this paper,by the choice of node root according to the dependence among attributes,a new TAN algorithm is proposed,CRTAN(Choose Root Tree Augmented Naive Bayes).Finally,we carry out experiments to evaluate the CRTAN and compare it with basic TAN and Naive Bayes.The experimental results show that the accuracy of the CRTAN is much higher than that of basic TAN and Naive Bayes on the datasets. |
| |
| Keywords: | classifie tree augmented naive bayes dependence |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
