摘要
本文旨在运用数学方法描述词汇与语篇长度的非线性关系,找出能够描写词汇与语篇长度关系的具有最佳拟合度的数学模型。主要研究英语篇际词汇覆盖率的分布特点及其分布的标准差,并运用描写词汇-语篇长度关系的最佳模型,推导出一个计算篇际词汇覆盖率的鲁棒理论数学模型,并计算其95%的置信区间。经检验,这一理论数学模型及其95%置信区间准确、可靠,可用于预测不同长度的真实语篇对的篇际词汇覆盖率的理论数值及其变化范围。
This paper intends to describe the non- linear relationship between vocabulary and text length via a mathematical method and tries to find out a mathematical model that has the best fit to the vocabulary-text length relationship. It mainly studies the distribution features of English inter-textual vocabulary coverage and its standard deviations, and employs the best model of vocabulary-text length relationship to infer a robust theoretical mathematical mode/, which can be used to calculate inter-textual vocabulary coverage. The 95% confidence intervals for the theoretical model are also calculated. It is tested that this theoretical mathematical model and its 95% confidence intervals are accurate and reliable and can be applied to predict the theoretical values of inter-textual vocabulary coverage for real text pairs of different lengths as well as their possible range of variation.
出处
《中国外语》
CSSCI
北大核心
2014年第6期53-61,共9页
Foreign Languages in China
基金
国家社科基金一般项目"英语篇际词汇覆盖模式与词汇教学研究"(编号:10BYY042)的研究成果
关键词
篇际词汇覆盖率
语料库
数学模型
95%置信区间
intertextual vocabulary coverage
corpus
mathematical model
95% confidence interval
