科技文本机器翻译句层错误分析
Error Analysis at the Sentence Level in Machine Translation of Sci-Tech Texts

机器翻译借助计算机对文本进行自动翻译，因翻译速度较快，现被普遍用于翻译实践中。由于中西文化和思维方式的不同，英汉语在句子层面存在明显的差异，这些差异对于机器翻译译文的影响是显而易见的。机器译文与人工译本相比，译文质量仍有一定的差距。因此，透析英汉语句子层面的差异，分析机器译文句层错误，有利于改进机器翻译的质量。本文以英汉语句层的差异为切入点，选取科技文本作为语料，将其机器译文与人工译本进行对比，从语法转换因子和逻辑关系两个方面归纳机器译文出现的句层错误，揭示机器翻译中应注意的几个问题，为提升机器翻译系统的形式规则提供语言基础，为译后编辑提升译文质量提供方向。
Machine translation (MT) is the process of translating texts from one language into another by computer. Due to its fast translation speed, MT has been widely used in translation practice. For different logical thinking and cultural background, differences can be found between English and Chinese at the sentence level. It’s obvious that these differences will have an impact on machine translation. There is a certain gap between the quality of machine translation and human translation, and errors can be found in the machine translation. Thus, mastering the differences between English and Chinese at the sentence level is very important for improving the quality and readability of machine translation. Based on the difference between English and Chinese at the sentence level, this paper intends to compare the machine translation and human translation of scientific and technological texts, and summarize the errors in machine translation at sentence level grammatically and logically, aiming to provide a language basis for improving the formal rules of machine translation system.