【作者】 吴溢；

【导师】 郑达华；

【作者基本信息】 浙江大学 ， 外国语言学及应用语言学， 2009， 硕士

【摘要】 机器翻译至今已有50多年的历史,从它诞生的第一天起,人们对它就褒贬不一。不能否认的是,尽管有其固有的问题,不断改进的机器翻译已经成为许多翻译工作者的翻译工具。机器翻译的方法规则一直都是研究者们热衷的一个课题,但是汉英机器翻译中的语言错误却并没得到太多的关注。本文中,两种目前最广泛使用的翻译工具----Systran和Google翻译了四个不同题材的文本,通过分析所得到的翻译文本,我们得到汉英机器翻译中最常见的语言错误。分析发现在汉英机器翻译中,歧义是最常见的错误,除了词的歧义之外,翻译连续的动词结构和连续的名词结构时出现歧义也是最多的。分析也同时发现,在整体上Google翻译软件的翻译质量要略高于Systran翻译软件的翻译质量,同时机器翻译在翻译复杂结构和多内容的文章时还有很大的不足。在文章最后作者提出了一些建议,重点解决在翻译连续动词和连续名词中出现的歧义问题。更多还原

【Abstract】 Machine translation (MT) has existed for more than 50 years. It received praises and criticism as well. One cannot deny that MT-much improved since then, is a useful tool for the human translator, although it inherently has many problems. Since it was first introduced researchers have showed great enthusiasm in developing the approaches to MT. However, one important issue which to our knowledge has not yet been widely investigated is the linguistic errors in MT. In the present study, we try to examine the linguistic errors through comparing the performance of Chinese-English machine translation in four different text genres which vary in their structures, using Systran Systems and Google Translator, which are widely used. Our results show that besides the lexical ambiguity, syntactic ambiguity involving serial verb construction and semantic ambiguity involving serial nouns are of most frequent ambiguities. Our results also show that on the whole Google translation tool did a relatively better job than Systran translation tool. In addition to the two findings, we find that MT is still limited in its ability to process certain text types, namely those with complex sentence structures, high amounts of pragmatic information and broad semantic domains. In the final part of the paper, the author came up with some suggestions to the disambiguity strategy with regard to the serial verb construction and serial noun construction.更多还原