平行文本中的对应词语序列的识别及应用研究

Corresponding Sequences in Parallel Texts

平行文本中存在大量的词语序列式的翻译对等,这些序列虽然不是翻译过程中最小的语言单位,但它们高频出现,体现了跨语言交际的意义实现机制和文本构成特点。本文基于交大平行语料库数据,重点介绍对应词语序列的识别和界定标准,以及对应词语序列在平行文本对齐和对比短语学等方面的相关应用研究,具体包括:双语双向的多级别对齐、词语搭配的对应、对应序列的形式与意义研究等。

There exist a huge not the smallest monosemous units meaning realization mechanism and number of translation equivalents of sequences in parallel texts. These sequences are in translation, but they occur highly frequently in the corpus and demonstrate the text composition features of cross-language communications. Based on the JiaoDa parallel corpus, this paper reports primarily the defining criteria of corresponding sequences, and introduces the related applied studies of corresponding sequences in such fields as alignment technology and phraseology in contrast, specifically including two-way alignments at multi-levels, corresponding lexical collocations, and the form and meaning of corresponding sequences.