摘要
理清文本互文性对于理顺文本间的多维关联对提高文本理解和翻译的准确性具有重要作用。以典籍文本《茶经》和《续茶经》为试验对象,使用向量空间模型(VSM)度量文本互文性,在此基础上提出了改进方法,即扩展项之间的相似度计算和采用序列模型,并给出了基于文本互文性度量的文本翻译索引方法。实验结果表明,扩展项之间的相似度计算效果不够理想,而采用序列模型则可取得较好的结果,可为原文本的准确理解和译文本的翻译提供有益参考。
Intertextuality clarification plays a significant role in clearing up the multilayer relationship among relevant texts,facilitating the process of text comprehension and translation.With two tea classics as case texts,a vector space model（VSM） in intertextuality calculation is presented.Then an improved method is proposed,i.e.extending the similarity calculation between items and using sequence model.Finally,a translation index method is produced.Experimental results show that the similarity calculation between items fails to provide satisfactory results,while the application of sequence model yields better results,which offers helpful clues to enhance the accuracy and consistency in both source text comprehension and target version translation.
出处
《计算机工程与设计》
CSCD
北大核心
2010年第15期3490-3493,共4页
Computer Engineering and Design
基金
国家自然科学基金项目(60673039)
大连理工大学2008人文社科研究基金项目(DUTHS2008320)
关键词
互文性
向量空间模型
序列模型
翻译索引
茶典籍
intertextuality
vector space model
series model
translation index
tea classics
