classifier_select_docs: Subsets a dataframe of documents based on a classifier.

In gidonc/durhamevp: ESRC/AHRC Victorian Election Violence Project R Code (Durham University)

Subsets a dataframe of documents based on a classifier.

Description

Subsets a dataframe of documents based on a classifier.

Usage

classifier_select_docs(
  classifier,
  new_docs,
  text_field = "description",
  return_logical = FALSE,
  logical_to_prob = FALSE,
  class_to_keep = 1,
  boolean = FALSE,
  xgb.cutpoint = 0.5,
  stem = FALSE,
  ...
)

Arguments

classifier	A classifier to perform the classification: either a naive bayes (quanteda) or xgboost (xgboost)
new_docs	A data frame or dfm containing the documents to classify
text_field	The field containing the text to classify
return_logical	Should the function return the subset of documents (FALSE) or a logical vector indicating the subset of document (TRUE).
logical_to_prob	return_logical == TRUE class probabilities can be returned instead of class categories (TRUE).
class_to_keep	The classifier class to keep
stem	stem words (in preprocessing)
...	other arguments to be passed to preprocess_corpus

Value

Either the subset of the docs_df which is classified as class_to_keep or a logical vector indicating this subset (depending on value of return_logical).

gidonc/durhamevp documentation built on April 8, 2022, 10:31 a.m.