count_words: Count words in child and adult speech

In rosemm/FrequencyFilter: The Frequency Filter

Description

Counts the number of occurrences of each word in the target child's speech and that of the child's primary interlocutor Optionally distinguish counts of eah word by part of speech by setting use.mor=TRUE (e.g. count "kiss" as a verb and "kiss" as a noun separately). It makes used of the part of speech tagging available in the MOR tier of CHAT transcribed files, but if desired one can further collapse those part of speech categories using the POS_regex argument, which will replace MOR part of speech with the user-specified labels. This is useful if, for example, one wants to count occurrences of words separately by broad part of speech category but not as fine-grained as MOR tags (e.g. count "kiss" as a noun separately from "kiss" as a verb, but collapse categories for child forms, family words and wordplay so that "beep" isn't counted separately for each.)

Usage

count_words(this.transcript, use.mor = TRUE, POS_regex = NULL,
  mor = "mor_word", orth = "orth_word", debug = FALSE)

Arguments

this.transcript	a dataframe with all words from the utterances from one transcript (or one timepoint, if transcripts from roughly the same age are being pooled together)
use.mor	whether or not to count occurrences separately by part of speech
POS_regex	if use.mor=TRUE, optionally collapse part of speech categories using regular expressions here
mor	the column that contains the MOR tier entries
orth	the column that contains the speaker tier entries
debug	turn on to display extra messages while the function runs, useful for identifying problem transcripts. Default is FALSE.

Value

A dataframe with word (from the speaker tier), gloss and POS (extracted from the mor tier), and counts from target child and target child's primary adult interlocutor. To count the words for several transcripts, count_words can be placed in side a for loop or do expression.

See Also

utts_to_words

Examples

## Not run: 
words.mor <- transcripts.cleaned %>%
group_by(language, corpus, child, age.mos, file, cha, speaker, name, role, utt.num) %>%
  do({
    utts_to_words(., debug = FALSE)
  }) %>%
  ungroup()

word.counts.mor <- words.mor %>%
group_by(language, corpus, child, age.mos, file, cha) %>%
  do({
    count_words(., use.mor = TRUE, POS_regex=NULL, debug = FALSE)
  })

## End(Not run)

rosemm/FrequencyFilter documentation built on May 29, 2019, 8:50 a.m.