Abstract: Tibetan-Chinese machine translation is different from Chinese-English machine translation. One important reason, Tibetan is more dependent on the role of function word like the case-auxiliary word in the sentences. Classes of case-auxiliary words are various and the use of them is extremely different. This paper is to analyze the Tibetan case-auxiliary words, integrate the semantic cues on the basis of Tibetan syntax tree, form the method of Tibetan-Chinese machine translation which is based on semantic cues. Through the experiment between the model of phrase and semantic, this method can be useful for Tibetan-Chinese machine translation.

Key words: semantic parsing, Tibetan syntax, Tibetan-Chinese machine translation

摘要： 藏汉机器翻译技术跟汉英机器翻译技术有所不同，其中，很重要的一个方面，藏语更依赖于格助词等虚词在句子中的作用，格助词种类繁多，用法差异很大。针对藏语格助词进行分析，在藏语短语句法树库的基础上，加入了藏语本体特征的语义信息，形成融合藏语语义信息的藏汉机器翻译方法。通过对比基于短语和句法的实验分析，该方法可以很好地应用于藏汉机器翻译系统。

关键词: 语义分析, 藏语句法, 藏汉机器翻译