基于本体的双语平行语料库的构建研究  

A Study on the Construction of the Ontology-based Chinese-English Parallel Corpus

作　　者：姚振军 郑旭红[2] 徐鹏涛[1] 

机构地区：[1]东北财经大学 [2]大连外国语大学

出　　处：《语言教育》2014年第1期29-36,共8页Language Education

摘　　要：无论是基于实例还是基于统计的机器翻译系统都对语料库的规模有着很高要求。目前语料库的构建基本上都是人工构建的,这无疑对语料库的构建规模有一定的限制。本文研究探讨引入本体技术,构建大规模双语平行语料库的可行性。基于本体的语料库构建主要有本体的实体抽取和推理归类处理两个方面。Both the example-based machine translation and the statistics-based machine translation have a high requirements for the size of the corpus. At present, most of the corpra are constructed with human power, which undoubtedly sets a limits on the size of corpus. This paper studies the feasibility of contructing large- scale Chinese-English parallel corpus with the introduction of ontology technology. The corpus construction based on the ontology technology consists of two aspects： extracting the individuals and reasoning classes.

关 键 词：机器翻译 语料库 本体 《道德经》 

分 类 号：H0[语言文字—语言学]