| 朴素贝叶斯分类器的误差估计 |
| |
| 引用本文： | 陈弋兰,;王鸣,;孙书诚.朴素贝叶斯分类器的误差估计[J].安徽机电学院学报,2008(4):48-51. |
| |
| 作者姓名： | 陈弋兰 ;王鸣 ;孙书诚 |
| |
| 作者单位： | [1]安徽建筑工业学院数理系,安徽合肥230601; [2]安徽工程科技学院安徽省电气传动与控制重点实验室,安徽芜湖241000 |
| |
| 摘 要： | 模式分类旨在依据识别对象特征的观察值将其归并至某个类别，贝叶斯分类决策可实现模式分类．给出一类遵从特征观察量独立性假设的贝叶斯分类器即朴素贝叶斯分类器设计．并给出朴素贝叶斯分类器的分类误差估计方法．理论分析与实验结果表明，朴素贝叶斯分类器设计方案可行，且其分类误差估计方法有效，可用于模式分类事务处理．
|
| 关 键 词： | 朴素贝叶斯分类器 误差估计 错误率 |
Error estimation in a naive bayes classifier |
| |
| Affiliation: | CHEN Yi-lan ,WANG Ming ,SUN Shu-cheng (1. Dept of Math. ＆ Phy. , Anhui University of Architecture, Hefei, 230601, Anhui; 2. Anhui Provincial Key Laboratory of Electric Transmission and Control, Anhui University of Technology and Science, Wuhu 241000, China) |
| |
| Abstract: | The pattern classification means to classify the object into a class by using naive Bayesian classification decision according to its eigenvalue. The Bayes classifier obeying the conditional independence assumption of the object＇s eigenvalue, namely the naive Bayes classifier, is designed, and the method of its classification error estimation is proposed. The theoretical analysis and the experimental result show that the design of the naive Bayes classifier is feasible, and the method of its classification error estimation is effective, for the pattern classification of objects. |
| |
| Keywords: | naive Bayes classifier error estimation error rate |
| 本文献已被 维普 等数据库收录！ |
|
