idiolect (version 1.0.1)

delta: Delta

Description

This function runs a Cosine Delta analysis (Smith and Aldridge 2011; Evert et al. 2017).

Usage

delta(
  q.data,
  k.data,
  tokens = "word",
  remove_punct = FALSE,
  remove_symbols = TRUE,
  remove_numbers = TRUE,
  lowercase = TRUE,
  n = 1,
  trim = TRUE,
  threshold = 150,
  features = FALSE,
  cores = NULL
)

Value

If features is set to FALSE then the output is a data frame containing the results of all comparisons between the Q texts and the K texts. If features is set to TRUE then the output is a list containing the results data frame and the vector of features used for the analysis.

Arguments

q.data

The questioned or disputed data, either as a corpus (the output of create_corpus()) or as a quanteda dfm (the output of vectorize()).

k.data

The known or undisputed data, either as a corpus (the output of create_corpus()) or as a quanteda dfm (the output of vectorize()).

tokens

The type of tokens to extract, either "word" (default) or "character".

remove_punct

A logical value. FALSE (default) keeps punctuation marks.

remove_symbols

A logical value. TRUE (default) removes symbols.

remove_numbers

A logical value. TRUE (default) removes numbers

lowercase

A logical value. TRUE (default) transforms all tokens to lower case.

n

The order or size of the n-grams being extracted. Default is 1.

trim

A logical value. If TRUE (default) then only the most frequent tokens are kept.

threshold

A numeric value indicating how many most frequent tokens to keep if trim = TRUE. The default is 150.

features

Logical with default FALSE. If TRUE, then the output will contain the features used.

cores

The number of cores to use for parallel processing (the default is one).

References

Evert, Stefan, Thomas Proisl, Fotis Jannidis, Isabella Reger, Steffen Pielström, Christof Schöch & Thorsten Vitt. 2017. Understanding and explaining Delta measures for authorship attribution. Digital Scholarship in the Humanities 32. ii4–ii16. https://doi.org/10.1093/llc/fqx023. Smith, Peter W H & W Aldridge. 2011. Improving Authorship Attribution: Optimizing Burrows’ Delta Method*. Journal of Quantitative Linguistics 18(1). 63–88. https://doi.org/10.1080/09296174.2011.533591.

Examples

Q <- enron.sample[c(5:6)]
K <- enron.sample[-c(5:6)]
delta(Q, K)