汉英机器翻译的英文词形态生成

MORPHOLOGIC GENERATION IN CHINESE-ENGLISH MACHINE TRANSLATION

在汉英机器翻译译文生成中，一个主要的问题是如何根据句子的上下文语境获取有关时态、语态、句式和主谓性、数、格等信息，生成具有正确单词形态的译文，如动词的过去式、过去分词、现在式形式；名词的所有相、复数形式；助动词生成以及冠同的生成等．本文提出一种基于SC文法的汉英机器翻译译文词形态生成算法，该方法通过设计一种生成导向的语言特征描述体系，采用译文生成和源文分析一体化的语言分析技术，使得译文生成能够充分利用源又分析过程中所用到的各种知识，准确地形成句子中各个成分的形态特征，并能有效地解决汉英机译译文生成中助动词和冠词的生成这一难题．

In the development of a Chinese-English machine translation system, one ofthe key problems is how to generate different morphologies of words(such as, the posses-sive and plural forms of noun, the past participial form of verb etc. ) according to currentcontext situation. In this paper, a SC grammar-based generation algorithm for Chinese-English machine translation is proposed. In which, by designing a generation-driven lin-guistic feature description scheme and taking a unified language analysis approach of incor-porating the generation with parsing, through using various knowledge which is producedand used in parsing, it is possible to generate proper morphologic feature of each compo-nent and to efficiently generate corresponding auxiliary verb and articles in sentence.