searchCorpus: Simple example function to extract data from the corpus

In jeroenclaes/tweetCorp: tweetCorp: A package to work with Twitter corpora

Description

There is a real risk of overflowing your computer's memory if you search for a frequent pattern in the files. If you specify an output directory, the results out kept out of memory. Linux/Mac users may want to use the command line tool 'grep' instead of this function, as it will be much faster. e.g. grep [PATTERN] corpus_directory/*csv > outputfile.csv. Information on POS and dependency labels:http://universaldependencies.org/format.html.

Usage

searchCorpus(pattern, corpus_directory, output_directory = NULL,
  field = c("tagged", "text"))

Arguments

pattern	word or regular expression
corpus_directory	directory where you keep the corpus file
output_directory	directory where you want the output files to go. Keep to NULL (default) if you want to collect the results in an R object rather than writing them out to files
field	one of 'tagged' or 'text'. If you want to search the parsed data, opt for 'tagged', otherwise, opt for 'text'

Value

Either a data.frame or nothing

jeroenclaes/tweetCorp documentation built on May 27, 2019, 4:50 a.m.