外语地名机器翻译中通专名区分技术研究  被引量:4

A Study of Distinguishing Technology in Translating Foreign Generic Names and Proper Names in Geography Through Machine

在线阅读下载全文

作  者:颜闻 毛曦[1] 钱赛男 马维军[1] 殷红梅[1] 刘德钦[1] YAN Wen;MAO Xi;QIAN Sainan;MA Weijun;YIN Hongmei;LIU Deqin(Chinese Academy of Surveying&Mapping,Beijing 100830,China;Zhejiang Academy of Surveying and Mapping,Hangzhou 311100,China)

机构地区:[1]中国测绘科学研究院,北京100830 [2]浙江省测绘科学技术研究院,浙江杭州311100

出  处:《测绘地理信息》2021年第3期118-121,共4页Journal of Geomatics

基  金:中国测绘学院研究院基本科研业务经费(7771802,7771721)。

摘  要:现有外语地名机器翻译中只有关于音译方面的研究,区分地名通名与专名的研究未开展,造成无法区分地名的通名与专名,外语地名翻译过程中无法正确选择音译与意译的翻译方式,外语地名翻译准确率不高。针对这一问题,提出一种外语地名通专名区分技术。该技术通过对地名词语构建结构树,并对树的叶子节点进行分类来完成通名与专名的区分,从而确定翻译方式。对英国地名进行区分,实验结果证明,该技术可以有效区分外语地名通专名,可辅助进一步的地名翻译,提高翻译精度。The current research on machine translation of geographical names focuses on transliteration while the study on distinction between generic and proper names of geographical names has not been carried out. It is impossible to distinguish the generic and proper names in geography and to select the translation methods of transliteration and free translation properly in the translation process, thus leading to the low accuracy of translating foreign place names. In view of this problem, a technique for distinguishing the generic and proper names is proposed. The technology first constructs a structural tree for the generic and proper names in geography, then distinguishes the generic and proper names through the classification of the leaf nodes of the tree, hence determines the translation method. The successful distinguishing of the place names in the United Kingdom proves that the technology can effectively distinguish foreign generic and proper place names and assist in the translation of further geographical names and improve the translation accuracy.

关 键 词:地名翻译 通名专名分类 结构树 互信息 

分 类 号:P281[天文地球—地图制图学与地理信息工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象