quanteda (version 1.1.1)

tokens_replace: Replace types in tokens object

Description

Substitute token types based on vectorized one-to-one matching. Since this function is created for lemmatization or user-defined stemming, it does not support multi-word features, or glob and regex patterns. Please use tokens_lookup with exclusive = FALSE for substitutions of more complex patterns.

Usage

tokens_replace(x, pattern, replacement = NULL, case_insensitive = TRUE,
  verbose = quanteda_options("verbose"))

Arguments

x

tokens object whose token elements will be replaced

pattern

a character vector or dictionary. See pattern for more details.

replacement

if pattern is a character vector, then replacement must be character vector of equal length, for a 1:1 match. If pattern is a dictionary, then replacement should not be used.

case_insensitive

ignore case when matching, if TRUE

verbose

print status messages if TRUE

Examples

# NOT RUN {
toks <- tokens(data_corpus_irishbudget2010)

# lemmatization
infle <- c("foci", "focus", "focused", "focuses", "focusing", "focussed", "focusses")
lemma <- rep("focus", length(infle))
toks2 <- tokens_replace(toks, infle, lemma)
kwic(toks2, "focus*")

# stemming
type <- types(toks)
stem <- char_wordstem(type, "porter")
toks3 <- tokens_replace(toks, type, stem, case_insensitive = FALSE)
identical(toks3, tokens_wordstem(toks, "porter"))
# }