R/predict.R

In gptzeror: Identify Text Written by Large Language Models using 'GPTZero'

#' Predict if a file was generated by AI
#'
#' @param file file to upload
#'
#' @return a tibble of sentence-level assessments
#' @export
#'
#' @examplesIf gptzeror::gptzero_has_key()
#' # Requires API Key
#' abstr <- 'Congressional district lines in many U.S. states are drawn by partisan
#' actors, raising concerns about gerrymandering. To separate the partisan effects
#' of redistricting from the effects of other factors including geography and
#' redistricting rules, we compare possible party compositions of the U.S. House
#' under the enacted plan to those under a set of alternative simulated plans
#' that serve as a non-partisan baseline. We find that partisan gerrymandering
#' is widespread in the 2020 redistricting cycle, but most of the electoral bias
#' it creates cancels at the national level, giving Republicans two additional
#' seats on average. Geography and redistricting rules separately contribute a
#' moderate pro-Republican bias. Finally, we find that partisan gerrymandering
#' reduces electoral competition and makes the partisan composition of the U.S. House
#' less responsive to shifts in the national vote.'
#'
#' abstr <- gsub('\\n', ' ', abstr) # remove the "\n"s
#'
#' fake_abstr <- 'The 2020 congressional redistricting process in the United States
#' has been marked by widespread partisan gerrymandering, which has led to concerns
#' about its impact on electoral outcomes and democratic representation. This paper
#' examines the effects of partisan gerrymandering on electoral competition in the
#' 2020 congressional elections. Using data from previous election cycles and
#' simulations of the 2020 elections, we find that while partisan gerrymandering
#' mostly cancels out at the national level, it reduces competition in individual
#' districts, making them less competitive and reducing voter choice. We also find
#' that the impact of gerrymandering on electoral outcomes varies significantly
#' by state and political party, suggesting that redistricting can have significant
#' implications for democratic representation and political power in the United
#' States. The findings of this study have important implications for policymakers,
#' as well as advocates of fair and democratic elections.'
#'
#' fake_abstr <- gsub('\\n', ' ', fake_abstr) # remove the "\n"s
#'
#' tf_abstr <- tempfile(fileext = '.txt')
#' cat(abstr, file = tf_abstr)
#'
#' tf_fake <- tempfile(fileext = '.txt')
#' cat(fake_abstr, file = tf_fake)
#'
#' gptzero_predict_file(tf_abstr)
#' gptzero_predict_file(tf_fake)
gptzero_predict_file <- function(file) {

  if (!is.character(file)) {
    cli::cli_abort('{.arg file} must be a character vector.')
  }
  if (length(file) != 1) {
    cli::cli_abort(
      c(
        '{.arg file} must be length 1.',
        'Iterate over {.fn gptzero_predict_file} for multiple requests.'
      )
    )
  }

  req <- httr2::request(base_url = api_url()) |>
    httr2::req_url_path_append('files') |>
    httr2::req_headers(
      `X-Api-Key` = gptzero_get_key(),
      accept = 'application/json'
    ) |>
    httr2::req_body_multipart(
      files = curl::form_file(file)
    )

  out <- req |>
    httr2::req_perform() |>
    httr2::resp_body_json()

  process_doc(out)
}

#' Predict if text was generated by AI
#'
#' @param text text to upload
#'
#' @return a tibble of sentence-level assessments
#' @export
#'
#' @examplesIf gptzeror::gptzero_has_key()
#' # Requires API Key
#' abstr <- 'Congressional district lines in many U.S. states are drawn by partisan
#' actors, raising concerns about gerrymandering. To separate the partisan effects
#' of redistricting from the effects of other factors including geography and
#' redistricting rules, we compare possible party compositions of the U.S. House
#' under the enacted plan to those under a set of alternative simulated plans
#' that serve as a non-partisan baseline. We find that partisan gerrymandering
#' is widespread in the 2020 redistricting cycle, but most of the electoral bias
#' it creates cancels at the national level, giving Republicans two additional
#' seats on average. Geography and redistricting rules separately contribute a
#' moderate pro-Republican bias. Finally, we find that partisan gerrymandering
#' reduces electoral competition and makes the partisan composition of the U.S. House
#' less responsive to shifts in the national vote.'
#'
#' abstr <- gsub('\\n', ' ', abstr) # remove the "\n"s
#'
#' gptzero_predict_text(abstr)
gptzero_predict_text <- function(text) {

  # check text input
  if (!is.character(text)) {
    cli::cli_abort('{.arg text} must be a character vector.')
  }
  if (length(text) != 1) {
    cli::cli_abort(
      c(
        '{.arg text} must be length 1.',
        'Iterate over {.fn gptzero_predict_text} for multiple requests.'
      )
    )
  }

  # formulate and perform request
  req <- httr2::request(base_url = api_url()) |>
    httr2::req_url_path_append('text') |>
    httr2::req_headers(
      `X-Api-Key` = gptzero_get_key(),
      accept = 'application/json'
    ) |>
    httr2::req_body_json(
      data = list(
        document = text
      )
    )

  out <- req |>
    httr2::req_perform() |>
    httr2::resp_body_json()

  process_doc(out)
}