摘要:
本文针对近期认知语言学研究中出现的定量转向,用具体案例介绍了一系列国际认知语言学研究中前沿的多元统计方法,主要涉及探索性统计分析(如聚类分析、多维尺度分析)、验证性统计分析(如逻辑斯蒂回归建模、混合效应回归建模)以及机器学习模型(如随机森林、朴素判别学习)三个方面。最后,本文还简要介绍了在这些数据统计分析中可资利用的统计软件包(R环境)。
ln line with the quantitative tum in Cognitive Linguistics , this paper introduces a series of advanced quantitative methods with ample examples. It illustrales multivariate methods of exploratory analysis (e. g. cluster analysis , multidimensional scaling analysis) , confirmalory analysis (e. g. logistic regression model , mixed-effects regression model) and machine learning model (e. g. random forest , naïve discriminative learning) . Moreover ,this paper briefly introduces statistical packages of those methods in the statistical environment R.