R/OBJECT_offensive.R

In sebastiansauer/pradadata: Data Sets for Practical Data Analysis

#' Text snippets, similar to tweets, in German language, partly offensive 
#' 
#' Short texts, similar to tweets, with a classification of offensive or not.
#' Two classifications are provided. `c1`  classifies as hateful or not, whereas
#' `c2` classifies the type of offensive content: abuse, insult, profanity or
#' other (not offensive).
#' 
#'
#' @docType data
#'
#' @usage data(offensive)
#'
#' @format A data frame containing 3 variable and approx. 5009 rows
#' \describe{
#'   \item{text}{Character. Tweet-like text}
#'   \item{c1}{Character. Offensive content or not?}
#'   \item{c2}{Character. Type of offensive content}

#' }
#' @source Wiegand, Michael. 2019b. GermEval-2018 Corpus (DE). heiDATA. https://doi.org/10.11588/DATA/0B5VML., \url{https://heidata.uni-heidelberg.de/dataset.xhtml?persistentId=doi:10.11588/data/0B5VML}. Licenced under CC-By-4.0 Int.

"offensive"