朝鲜语特殊短语结构的语言特征分析及自动提取

朝鲜语中存在大量特殊短语结构,因此在朝汉翻译中,如何准确翻译这些特殊短语显得尤为重要,此举有利于提高机器翻译的精度与效率。本文基于韩国"世宗计划"标注语料库,通过对特殊短语结构进行语言特征分析,构建规则库,以迭代方式提取特殊短语结构及其分布,并以中心词为""的特殊短语为例,进行自动提取实验,取得满意的效果。

Structure Characteristic Analysis and Automatic Extraction for Korean Special Phrase

There are a large number of special phrase structures in Korean, so it is especially important to accurately translate these special phrases in the process of Korean-Chinese translation, which is helpful to promote the accuracy and efficiency of machine translation. This article was based on the tagged corpus made by Korea''s Sejong plan. Through analyzing the language features of special phrases, a rule base was constructed, and the special phrase structure and its distribution were extracted with iterative method. The headword of "-따르다" in special phrases was used as an example to conduct an automatic extraction experiment and achieved satisfactory results.