corpustools (version 0.5.1)

plot.vocabularyComparison: visualize vocabularyComparison

Description

visualize vocabularyComparison

Usage

# S3 method for vocabularyComparison
plot(
  x,
  n = 25,
  mode = c("both", "ratio_x", "ratio_y"),
  balance = T,
  size = c("chi2", "freq", "ratio"),
  ...
)

Arguments

x

a vocabularyComparison object, created with the compare_corpus or compare_subset method

n

the number of words in the plot

mode

use "both" to plot both overrepresented and underrepresented words using the plot_words function. Whether a term is under- or overrepresented is indicated on the x-axis, which shows the log ratios (negative is underrepresented, positive is overrepresented). Use "ratio_x" or "ratio_y" to only plot overrepresented or underrepresented words using dtm_wordcloud

balance

if TRUE, get an equal amount of terms on the left (underrepresented) and right (overrepresented) side. If FALSE, the top chi words are used, regardless of ratio.

size

use "freq", "chi2" or "ratio" for determining the size of words

...

additional arguments passed to plot_words ("both" mode) or dtm_wordcloud (ratio modes)

Examples

## as example, compare SOTU paragraphs about taxes to rest
tc = create_tcorpus(sotu_texts[1:100,], doc_column = 'id')
comp = compare_subset(tc, 'token', query_x = 'tax*')

# \donttest{
plot(comp, balance=TRUE)
plot(comp, mode = 'ratio_x')
plot(comp, mode = 'ratio_y')
# }