| 文本倾向识别的置信度估计 |
| |
| 引用本文： | 郭燕慧,王小捷,钟义信.文本倾向识别的置信度估计[J].计算机工程与应用,2002,38(1):19-21. |
| |
| 作者姓名： | 郭燕慧 王小捷 钟义信 |
| |
| 作者单位： | 北京邮电大学智能研究中心,北京,100876 |
| |
| 基金项目： | 国家863高技术研究发展计划应急项目(编号:863-301-04-02-A) |
| |
| 摘 要： | 由于置信度模型可以有效地判断观测数据与文本倾向模板之间的匹配程度,因此可以用在文本倾向识别中,有效地定位识别结果,从而提高系统的识别率和稳健性。该文讨论了文本倾向识别中的置信度的基本原理,介绍了假设检验和区间估计在文本倾向识别中的应用。
|
| 关 键 词： | 置信度 文本倾向识别 假设检验 区间估计 |
| 文章编号： | 1002-8331-(2002)01-0019-03 |
| 修稿时间： | 2001年9月1日 |
The Theory and Applications of Confidence Measures in Text Tendency Recognition |
| |
| Guo Yanhui,Wang Xiaojie Zhong Yixin.The Theory and Applications of Confidence Measures in Text Tendency Recognition[J].Computer Engineering and Applications,2002,38(1):19-21. |
| |
| Authors: | Guo Yanhui Wang Xiaojie Zhong Yixin |
| |
| Abstract: | As confidence measures can well judge whether an observation matches a model,they can be used to locate the result of recognition in text tendency perfectly,thus improving accuracy and robustness of text tendency recognition.In this paper,the theory of confidence measures is introduced in the context of text tendency recognition,the applications of hypothesis testing and interval estimating are described. |
| |
| Keywords: | confidence measures text tendency recognition hypothesis testing interval estimating |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
