clean_text: Clean Input Text

In nlanderson9/languagePredictR: Predict Outcomes from Natural Language

Description

This function cleans text by:
-Setting all text to lowercase
-Removing non-ASCII characters
-Expanding contractions ("don't" –> "do not")
-Removing punctuation
-Removing symbols (if replaceSymbol is FALSE)
-Removing numbers (if replaceNumber is FALSE)

Usage

clean_text(
  inputText,
  replaceSymbol = FALSE,
  replaceNumber = FALSE,
  removeStopwords = FALSE
)

Arguments

inputText	A character string or vector of character strings
replaceSymbol	If TRUE, symbols are replaced with their equivalent (e.g. "@" becomes "at"). Defaults to FALSE.
replaceNumber	If TRUE, numbers are replaced with their equivalent (e.g. "20" becomes "twenty", "3rd" becomes "third"). Defaults to FALSE.
removeStopwords	If TRUE, stopwords are removed (see [stopwords()])

Value

A character string (or vector of character strings) with cleaned text.

Examples

myString = "He gave his last $10 to Sally's sister because she's nice."

cleanText = clean_text(myString)
# "he gave his last to sally sister because she is nice"

cleanText = clean_text(myString, replaceNumber = TRUE)
# "he gave his last ten to sally sister because she is nice"

cleanText = clean_text(myString, replaceSymbol = TRUE)
# "he gave his last dollar to sally sister because she is nice"

nlanderson9/languagePredictR documentation built on June 10, 2021, 11 a.m.