make.frequency.list {stylo}

Make List of the Most Frequent Elements (e.g. Words)

Description

Function for generating a frequency list of words or other (linguistic) features. It basically counts the elements of a vector and returns a vector of these elements in descending order of frequency.

Usage

make.frequency.list(data, value = FALSE, head = NULL, relative = TRUE)

Arguments

data

either a vector of elements (e.g. words, letter n-grams), or an object of a class stylo.corpus as produced by the function load.corpus.and.parse.

value

if this function is switched on, not only the most frequent elements are returned, but also their frequencies. Default: FALSE.

head

this option is meant to limit the number of the most frequent features to be returned. Default value is NULL, which means that the entire range of frequent and unfrequent features is returned.

relative

if you've switched on the option value (see above), you might want to convert your frequencies into relative frequencies, i.e. the counted occurrences divided by the length of the input vector – in a vast majority of cases you should use it, in order to neutralize different sample sizes. Default: TRUE.

Value

The function returns a vector of features (usually, words) in a descending order of their frequency. Alternatively, when the option value is set TRUE, it returns a vector of frequencies instead, and the features themselves might be accessed using the generic names function.

Author(s)

Maciej Eder

See Also

load.corpus.and.parse, make.table.of.frequencies

Examples

# assume there is a text:
text = "Mr. Sherlock Holmes, who was usually very late in the mornings, 
       save upon those not infrequent occasions when he was up all night, 
       was seated at the breakfast table. I stood upon the hearth-rug and 
       picked up the stick which our visitor had left behind him the night 
       before. It was a fine, thick piece of wood, bulbous-headed, of the 
       sort which is known as a \"Penang lawyer.\""

# this text can be converted into vector of words:
words = txt.to.words(text)

# an avanced tokenizer is available via the function 'txt.to.words.ext':
words2 = txt.to.words.ext(text, corpus.lang = "English.all")

# a frequency list (just words):
make.frequency.list(words)
make.frequency.list(words2)

# a frequency list with the numeric values
make.frequency.list(words2, value = TRUE)

## Not run: 
# #####################################
# using the function with large text collections

# first, load and pre-process a corpus from 3 text files:
dataset = load.corpus.and.parse(files = c("1.txt", "2.txt", "3.txt"))
#       
# then, return 100 the most frequent words of the entire corpus:
make.frequency.list(dataset, head = 100)

## End(Not run)