Frequency of Dictionary Terms in a Corpus

Calculates the absolute or relative frequency of dictionary terms in a given corpus.

Usage

countKeywords(corpus, 
              dic, 
              group.var=NULL, 
              rel.freq=FALSE, 
              case_insensitive=TRUE,
              quietly=FALSE)

Arguments

corpus

A quanteda corpus containing texts.

dic

A quanteda dictionary object containing the terms to be searched.

group.var

The variable in the corpus metadata to decompose the results. The default is NULL (all groups will be considered without desaggregation).

rel.freq

Establishes whether the function calculates the relative frequency. The default is FALSE.

case_insensitive

Indicates if the search will be insensitive to lower and uppercase letters. The default is TRUE.

quietly

Logical. Indicates if the function hides the progress bar or not. The default is FALSE.

Details

The function countKeywords calculates number of times each term in a dictionary appears in a corpus. It allows both absolute and relative frequencies. Grouping variables contained in corpus metadata can also be employed to disaggregate the result by categories.

Value

A data.frame object containing the groups (all if the group.var argument is not specified), the categories of the dictionary, the keywords composing each category or coding level, and the frequency.

Examples

if (FALSE) {
# Create a corpus object
library(quanteda)
cb <- corpus(spa.inaugural)

# use the dic.pol.es dictionary
dic <- dic.pol.es

# Generates the frequencies
d <- countKeywords(cb, 
                   dic)
                   
# Generates the relative frequencies 
# by President
d <- countKeywords(cb, 
                   dic,
                   group.var="President",
                   rel.freq=TRUE)
}