text_win_rqa: Window Recurrence quantification analysis ib text

In crqanlp: Cross-Recurrence Quantification Analysis for Dynamic Natural Language Processing

Description

It computes recurrence of text in overlapping windows over a number of delays smaller than the size of the window.

Usage

text_win_rqa(rsrc, typ = 'file', winsz = 10, wshft = 10, 
removeStopwords = F,embed = 1,tw = 1,limit = -1,shuffle = F)

Arguments

rsrc	Location of file or resource, or string literal
typ	A flag indicating the type of resource file in input: typ = "file" (it's a file name); typ = "ulr" (it's a url, and the file gets downloaded); typ = "string" or "raw_chars" (it's a literal string); typ = "tibble" (it's a text formatted as tidytext by tibble)
winsz	The size of the window
wshft	Interval by which the window is moved.
removeStopwords	A boolean: TRUE (remove stop words) - FALSE (it retains them)
embed	The number of embedding dimension for phase-reconstruction, i.e., the lag intervals.
tw	The Theiler window parameter
limit	A scalar indicating how much text should be considered
shuffle	A boolean if TRUE, it randomly shuffles the order of the text for surrogate analyses.

Details

A wrapper to the 'wincrqa()' function that runs windowed recurrence quantification analysis on text. This function also calls 'get_text_series()' to simplify the text in case such simplification was not done before inputting the text.

Value

It returns a matrix where the rows are the different windows explored, and the columns are the recurrence measures observed in that particular window. Refer to text_win_rqa for the values returned.

Author(s)

Rick Dale ([e-mail])

Examples

library(gutenbergr)
## let's get Alice's Adventures in Wonderland by Carroll
# gutenberg_works(author == "Carroll, Lewis") 
rsrc = gutenberg_download(11) ## take the text
win_rqa_res = text_win_rqa(rsrc, typ = "tibble", wshft = 150, winsz = 200, 
                       removeStopwords = FALSE, embed = 1, tw = 1,limit = -1,shuffle = FALSE)
win_rqa_res = as.data.frame(win_rqa_res$crqwin)

plot(win_rqa_res$window, win_rqa_res$RR)
abline(lm(win_rqa_res$RR ~ win_rqa_res$window), lwd = 3, col = "red")

crqanlp documentation built on May 2, 2019, 1:09 p.m.