loglikelihood: log-likelihood calculation derived from...

In cokelly/collocateR: Processes collocates for keywords in context and calculate collocate significance

Description

log-likelihood calculation derived from http://ucrel.lancs.ac.uk/llwizard.html

Usage

loglikelihood(doc = doc, keyword = keyword, window = 6, ngram = 1,
  remove_stopwords = TRUE, min_count = 2, span = "both")

Arguments

doc	Acharacter vector or list of character vectors
keyword	A key word or phrase to test
window	The number of context words to be displayed around the keyword Default 6
ngram	The size of phrases the frequencies of which we are to test (so, unigram = 1, bigram = 2, trigram = 3 etc)
remove_stopwords	Remove stopwords, derived from Quanteda's list
min_count	Collocates that occur fewer times than floor will be removed
span	Whether to include a window's width of words to the left of the keyword, to the right or on both sides

cokelly/collocateR documentation built on May 13, 2019, 8:49 p.m.