| 中文单字国名简称的自动识别 |
| |
| 引用本文： | 李斌,方芳. 中文单字国名简称的自动识别[J]. 计算机工程与应用, 2006, 42(28): 167-169,176 |
| |
| 作者姓名： | 李斌 方芳 |
| |
| 作者单位： | 南京师范大学文学院,南京,210097;南京师范大学文学院,南京,210097 |
| |
| 摘 要： | 单字地名简称识别是中文专名识别的重要组成部分,其中单字国名简称又占到了78.43%。但就现有的词性标注系统来看,还不能很好地对其进行识别。文章提出了一个基于规则的识别方法,以分词文本为底本,通过两次扫描,第一次扫描建立基于语篇的临时国名词表,第二次扫描利用上下文特征词等评分机制,从分词碎片中召回单字国名简称。在封闭测试和开放测试中,都取得了较好的实验结果,以人工分词文本为输入底本,调和平均值分别达到了96.33%和94.96%。
|
| 关 键 词： | 单字地名 单字国名简称 评分机制 临时词表 |
| 文章编号： | 1002-8331（2006）28-0167-03 |
| 收稿时间： | 2005-12-01 |
| 修稿时间： | 2005-12-01 |
Single Chinese Character Country Name Recognition |
| |
| LI Bin,FANG Fang. Single Chinese Character Country Name Recognition[J]. Computer Engineering and Applications, 2006, 42(28): 167-169,176 |
| |
| Authors: | LI Bin FANG Fang |
| |
| Affiliation: | Nanjing Normal University,Nanjing 210097 |
| |
| Abstract: | Single Character(shortened form) Country Name(SCCN),is a kind of shorted form of a Country name,which composed of one Chinese Character,such as “英”(ying1,England)and “美”(mei3,America).The SCCN recognition is a part of the task of Chinese Named Entity Recognition(NER).This paper investigates the usage of SCCNs in real text, then builds a modal for recognition based on rules,using temporary wordlist and contextual information as main resources.Using the manual segmented text as input,F-score of our method achieved 96.33% in the close test and 94.96% in the open test. |
| |
| Keywords: | Single Character Location Name Single Character Country Name evaluation nechanism temporary wordlist |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
