finnsurveytext (version 2.1.1)

fst_ngrams_table: Make Top N-grams Table

Description

Creates a table of the most frequently-occurring n-grams within the data. Optionally, weights can be provided either through a `weight` column in the formatted data, or from a `svydesign` object with the raw (preformatted) data.

Usage

fst_ngrams_table(
  data,
  number = 10,
  ngrams = 1,
  norm = NULL,
  pos_filter = NULL,
  strict = TRUE,
  use_svydesign_weights = FALSE,
  id = "",
  svydesign = NULL,
  use_column_weights = FALSE
)

Value

A table of the most frequently occurring n-grams in the data.

Arguments

data

A dataframe of text in CoNLL-U format, with optional additional columns.

number

The number of n-grams to return, default is `10`.

ngrams

The type of n-grams to return, default is `1`.

norm

The method for normalising the data. Valid settings are `"number_words"` (the number of words in the responses), `"number_resp"` (the number of responses), or `NULL` (raw count returned, default, also used when weights are applied).

pos_filter

List of UPOS tags for inclusion, default is `NULL` which means all word types included.

strict

Whether to strictly cut-off at `number` (ties are alphabetically ordered), default is `TRUE`.

use_svydesign_weights

Option to weight words in the table using weights from a `svydesign` containing the raw data, default is `FALSE`

id

ID column from raw data, required if `use_svydesign_weights = TRUE` and must match the `docid` in formatted `data`.

svydesign

A `svydesign` which contains the raw data and weights, required if `use_svydesign_weights = TRUE`.

use_column_weights

Option to weight words in the table using weights from formatted data which includes addition `weight` column, default is `FALSE`

Examples

pf <- c("NOUN", "VERB", "ADJ", "ADV")
pf2 <- "NOUN, VERB, ADJ, ADV"
fst_ngrams_table(fst_child, norm = NULL)
fst_ngrams_table(fst_child, ngrams = 2, norm = "number_resp")
fst_ngrams_table(fst_child, ngrams = 2, pos_filter = pf)
fst_ngrams_table(fst_child, ngrams = 2, pos_filter = pf2)
c2 <- fst_child_2
s <- survey::svydesign(id=~1, weights= ~paino, data = child)
i <- 'fsd_id'
fst_ngrams_table(c2, use_svydesign_weights = TRUE, svydesign = s, id = i)
fst_ngrams_table(fst_child, use_column_weights = TRUE, ngrams = 3)