当前位置: X-MOL 学术Journal of Quantitative Linguistics › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Numerical Assessment of Orthographic Neighbourhood Size Fluctuation in Writing Using Fractal Dimension Analysis
Journal of Quantitative Linguistics ( IF 0.761 ) Pub Date : 2019-11-25 , DOI: 10.1080/09296174.2019.1694360
Rex Taibu 1 , Eric Cheung 2 , Weier Ye 3 , Sunil Dehipawala 1 , Vazgen Shekoyan 1 , George Tremberger 1 , Tak Cheung 1
Affiliation  

ABSTRACT

The orthographic size of a targeted word, the number of new words that can be generated from a targeted word by exchanging a single letter, offers a research window where words can be transformed into numerical values. The CLEARPOND technology from Northwestern University was used for the transformation. A writing can then be modelled as a time series where the fluctuation can be further described using fractal dimension analysis. This project used the Higuchi fractal method for the computation of the fractal dimensions of time series. The proof of concept was conducted using writing examples which include Astronomy writing and English writing, the responses of Trump and Clinton in a Presidential election debate, and song lyrics. The results suggested that a high fractal dimension has an association with a high-demand cognitive task. The use of fractal dimension analysis as a writing assessment tool is discussed with relationship to the current lexical diversity computation technology.



中文翻译:

使用分形维数分析对写作中正字法邻域大小波动的数值评估

摘要

目标词的正字法大小,即通过交换单个字母可以从目标词中生成的新词的数量,提供了一个研究窗口,可以将单词转换为数值。改造使用了西北大学的CLEARPOND技术。然后可以将写作建模为时间序列,其中可以使用分形维数分析进一步描述波动。本项目使用樋口分形方法计算时间序列的分形维数。概念验证是使用写作示例进行的,其中包括天文学写作和英语写作、特朗普和克林顿在总统选举辩论中的反应以及歌词。结果表明,高分形维数与高要求的认知任务有关。

更新日期:2019-11-25
down
wechat
bug