【作者】 马红妹； 齐璇； 王挺； 陈火旺；

【Author】 MA Hongmei QlXuan WANG Ting CHEN Huowang(School of Computer, National University of Defence Technology, Changsha 410073, China)

【机构】 国防科技大学计算机学院；

【摘要】 汉语中主语省略现象十分普遍,汉语主语省略句的处理对于汉英机器翻译十分重要,它需要基于篇章上下文语境进行分析,包括省略主语识别和省略主语恢复。本文首先介绍了汉英机译系统ICENT的句法语义分析,然后建立了汉语篇章上下文语境模型,制定了主语省略恢复规则,给出了基于汉语篇章上下文语境应用主语省略恢复规则恢复主语省略的算法,最后对小学语文课本实际语料进行了实验.更多还原

【Abstract】 Zero-anaphoric subject of Chinese sentence is very prevalent It is important to process zero-anaphoric subjects of Chinese sentences in Chinese-English machine translation. The process needs to analyze Chinese text based on the knowledge of linguistic context, it includes zero-anaphoric subject recognizing and zero-anaphoric subject recovering. In this paper, syntactic analysis and semantic analysis of Chinese-English machine translation system ICENT are introduced firstly. Then the model of linguistic context of Chinese text is constructed, and zero-anaphoric subject recovering rules are designed With the linguistic context of Chinese text, Zero-anaphoric subject recovering algorithm (ZSRA) is proposed. Finally, experiment is done using the Chinese textbooks of elementary school.更多还原