面向汉英机器翻译的专利文献小句变换研究

专利文献的自动翻译是机器翻译的一个重要应用领域，复杂长句的翻译是汉英机器翻译的难点。本研究期望找出汉英复杂长句中小句变换的形式化转换规则。汉语复杂长句中会包含多个小句，这些小句都是独立存在的，但翻译成英语时，一般只有一个核一心小句，其他小句都变换成doing、todo、从句或短语等其它形式。文中以1300句汉英双语专利文献语料为研究对象，对汉语中的小句翻译为英语的变换情况进行分类研究，从小句句间关系、共享关系的角度出发，描述激活特征，并按五种变换方式分类，提出了十二条变换规则，小规模语料实验结果证明规则可行有效。下一步工作需要扩充研究语料，对语料进行更深入的挖掘和分析，在更大规模语料中验证规则的实用性。

A Chinese-English MT-Oriented Study on Small Sentence Pattern Transformation of Long Patent Sentence

Automatic translation of the patent document is an important field of machine translation, and the translation of long patent sen- tence is the difficulty in Chinese-English machine translation research. A complex sentence contains more than one small sentence, which is independent in Chinese. There is only one core sentence when translated into English from Chinese. Other sentences are transformed in- to clauses or non-defining phrases. It expected to find the transformation rules that how Chinese small sentence translated into English. Study 1300 Chinese-English bilingual sentences, from the standpoint of small sentence pattern transformation, according to semantic rela- tionship and sharing relationship among small sentences,deseribe the activation characteristics, sum up 12 formal sentence analysis rules and transformation rules. The small scale experiment results indicate these rules are effective. The next step needs to expand the research corpus and verify the practicality of these rules in larger corpus.