stabilize.estimate: Get stabilization estimate

Description Usage Arguments Value See Also Examples

View source: R/stabilize_estimate.R

Description

stabilize.estimate is used to find the minimum size of text needed to get a stable entropy rate.

Usage

1
2
stabilize.estimate(text, step.size, max.length = length(text),
  every.word = 10, method = "downsample", rate = 5, verbose = TRUE)

Arguments

text

Character vector. Preferably this should be read with read.file to take care of preprocessing. Note that internally this uses a regular expression for the matching so in non-ASCII texts it might be unstable. In those cases it is recommended to normalize the text beforehand.

step.size

A numeric scalar. Pre-specifies the step sizes at which entropy rates are computed. Warning: smaller values increase the computation time.

max.length

A numeric scalar. How many tokens of the text should be read? Defaults to the length of the text.

every.word

A numeric scalar. "every.word=1" specifies that each word token should be used for estimation. To speed up processing only every 2nd, 3rd, nth word token could be used. Hence, every.word can be assigned any integer between 1 and the step size.

verbose

Boolean. Whether or not to display progress.

downsampling.rate

A numeric scalar.

Value

stabilize.estimate returns an object of class Stabilize.

See Also

Stabilize, StabilizeCriterion

Examples

1
2
data(deuparl)
ce <- stabilize.estimate(deuparl, step.size = 1000, max.length = 10000, every.word = 10, method = "downsample", rate = 5, verbose = TRUE)

dimalik/Hrate documentation built on May 24, 2019, 4:01 a.m.