| 面向汉维机器翻译的相关转换及匹配规则设计 |
| |
| 引用本文： | 贾春兰,吐尔根·依布拉音.面向汉维机器翻译的相关转换及匹配规则设计[J].计算机应用与软件,2012(3):216-218,238. |
| |
| 作者姓名： | 贾春兰 吐尔根·依布拉音 |
| |
| 作者单位： | 新疆大学信息科学与工程学院 |
| |
| 摘 要： | 汉语和维吾尔语是在句法结构和语序上差异较大的两种语言。对于一个完备的汉维机器翻译系统而言,进行源语言的分析和目标语言时态、语态的准确表达是有必要的。针对统计机器翻译模型中所包含的句法、语义成分较低导致的准确率及语序问题,通过建立相关转换及匹配规则,以期用于机器翻译的混合方法之中来提高翻译系统的工作性能。
|
| 关 键 词： | 时态 语态 语义 机器翻译 转换 匹配规则 |
DESIGNING RELATED TRANSFORMATION AND MATCHING RULES FOR CHINESE-UYGHUR MACHINE TRANSLATION |
| |
| Affiliation: | Jia Chunlan Turgun Ibrahim(School of Information Science and Engineering,Xinjiang University,Urumqi 830046,Xinjiang,China) |
| |
| Abstract: | Chinese and Uyghur belong to two languages with great difference in terms of syntax structure and word order.As for a complete Chinese-Uyghur machine translation system,it is necessary to analyse the source language and to express the tense and voice of the target language accurately.In light of the problems of the accuracy rate and the word order caused by relatively lower syntactic and semantic ingredients contained in statistical machine translation model,in this paper we try to establish a set of related transformation and matching rules with the hope that it is to be applied to hybrid method of machine translation to improve the working performance of translation system. |
| |
| Keywords: | Tense Voice Semantics Machine translation Transformation Matching rules |
| 本文献已被 CNKI 等数据库收录！ |
|
