| 统计与规则相结合的藏文人名自动识别研究 |
| |
| 引用本文： | 窦嵘,加羊吉,黄伟.统计与规则相结合的藏文人名自动识别研究[J].长春工程学院学报(自然科学版),2010,11(2):113-115. |
| |
| 作者姓名： | 窦嵘 加羊吉 黄伟 |
| |
| 作者单位： | 甘肃同元信息系统技术有限责任公司;西北民族大学,中国民族语言文字信息技术重点实验室,兰州,730030;周口师范学院,计算机科学技术系,周口,466001 |
| |
| 摘 要： | 提出了一种基于统计与规则相结合的藏文人名自动识别方法.该方法充分挖掘了人名与上下文信息的关联程度,引入互信息对关联程度进行了定量的描述.根据大规模语料(30.2 M文本语料)的统计数据所建立的规则极大地提高了识别算法的准确率;在处理汉族音译人名情况时,采用姓氏驱动的方法,很好地解决了藏文文本中汉族人名的识别问题.实验结果表明,该方法具有较高的准确率和召回率.
|
| 关 键 词： | 藏文人名识别 互信息 格助词 人名词典 |
Automatic recognition of tibetan name with the combination of statistics and regular |
| |
| DOU Rong,JIA Yang-ji,HUANG Wei.Automatic recognition of tibetan name with the combination of statistics and regular[J].Journal of Changchun Institute of Technology(Natural Science Edition),2010,11(2):113-115. |
| |
| Authors: | DOU Rong JIA Yang-ji HUANG Wei |
| |
| Institution: | DOU Rong,etc.(Gansu Tongyuan InformationalTechnologies Co,Lanzhou 30030,China) |
| |
| Abstract: | This paper proposes a recognition method of Tibetan name based on the combination of statistics and regular.The method excavates the association degree of names and the textual information,and then introduces the mutual information to describe the association degree in statistic;statistic the rules which is built from the large-scale corpus(30.2 M text corpus) to improve the accuracy of recognition algorithm;use name-driven method to deal with the pronunciation translating of Han names,in order to get a goo... |
| |
| Keywords: | Tibetan names recognition mutual information case-auxiliary word name lexicon |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
