聚类分析在外国语言学研究中的应用

摘要

A survey indicates that most foreign language researchers across China lack an understanding of cluster analysis and cannot apply it in linguistic studies .Based on the analysis of corpus data , the authors of this paper hold that cluster analysis can be used in the following areas : lexical semantics , syntactic pattern studies , dialectology , stylistic , social and functional variation of language . Besides being a statistical technique ,cluster analysis also has methodological implications : it can help researchers systematically discover variables relevant for subject matter ; and it is a tool of generating high quality hypotheses in a data - driven way . It is also pointed out that cluster analysis can serve as an important quantitative research tool for systemic functional linguistics .%调查显示，我国大部分外国语言学研究者对聚类分析既缺乏了解也不会应用。本文展示了聚类分析和语料库结合可以应用于外国语言学的很多研究领域：词汇语义、句法型式，以及语言的地理变异、文体变异、社会变异和功能变异。聚类分析不仅是一种统计分析技术，而且具有研究方法论意义，即可以以数据驱动的方式系统性的发现研究变量和生成高质量的假设。还特别指出聚类分析可以作为系统功能语言学的定量化研究的一种重要工具。