【摘要】 语料库语言学需要从大规模文本提取语言特征,通过量化分析研究语言规律。现有语料库工具过于注重索引和检索功能,无法开展涉及复杂统计的多因素分析。通过3个基于语料库的研究实例,探讨编程语言Perl和R在研究方法层面的应用。结果表明,Perl和R能够处理大规模文本,进行多变量统计与可视化分析,可以弥补现有语料库软件的不足,帮助研究者分析数据与验证假设,为后续定性研究奠定基础。更多还原

【Abstract】 Corpus linguistics aims to find language patterns based on linguistic features extracted from large-scale texts.However,current corpus tools are dedicated to developing concordance and search functions while lack of functions to perform multivariate statistical analysis.This paper illustrates with three case studies how programming languages such as Perl & R can be used in corpus-based linguistic studies.It is found that Perl can extract linguistic features from texts and organize them in formats that are amenable to statistical analysis in R.When combined,these two kinds of software can help researchers explore the linguistic data and validate search hypothesis in a more flexible way and complement the functions of ready-made corpus tools.更多还原