基于语料库的汉语语块分类研究

A Corpus-based Study on Chinese Language Chunks Classification

文章采取基于语料库的研究方法,从语言使用的概率性特点入手,考察现代汉语系统中相关词语序列的使用频率以及相互搭配强度,对语块进行识别和判定。在对外汉语教学视角下,从语义、语法、语用等角度对语块进行多维度的分类,并从语言学视角对汉语语块的特点进行分析。文章认为:高频率、高互信息值的搭配类语块在语言使用中的整体提取性应该得到重视,它们是语块的重要组成部分。特别是跨层类组构语块,更是语料库语言学视角下语言系统中的一个重要聚类。

Based on the Corpus research method and starting with the probabilistic characteristics of language use, this paper explores the usage frequency of relevant words sequence and mutual collocation strength under modern Chinese system. From the perspective of teaching Chinese as a foreign language, the author classifies the lexical chunks in terms of semantics, syntax and pragmatics and other multi-dimensional angles. Based on which, this paper conducts the analysis on the characteristics of lexical chunks frorri the angles of linguistics and psychology respectively. This paper concludes that the overall extracting of collocation chunks featuring high frequency and high mutual-information value, as an important composition of chunks, should be attached great importance in language use. Cross-layer chunks, in particular, are more important clusters in the language system of Corpus linguistics perspective.