Visualize TF-IDF

Visualize term-frequency, inverse-document-frequency.

visualize_tfidf_at(..., token = NULL, doc = NULL, facet = doc,
  color = facet, num_top = 10, scale_manual_base = default_scale_manual(),
  scale_manual_params = list(), labs_base = default_labs(),
  labs_params = list(title = "Highest TF-IDF Words"),
  theme_base = default_theme(axis.text.y = ggplot2::element_text(angle = 30,
  hjust = 1), panel.grid.major.y = ggplot2::element_blank()),
  theme_params = list(), facet_base = default_facet(facet),
  facet_params = list())

visualize_tfidf(..., token, doc, facet, color)

Arguments

...	dots. Parameters to pass directly to visualize_time().
token	bare for NSE; character for SE. Name of column in data to use as unit for counting. Probably something like 'word' or 'bigram'.
doc	bare for NSE; character for SE. Name of column in data to use as document in tidytext::bind_tf_idf().
facet	bare for NSE; character for SE. Name of column in data to use for facetting.
color	bare for NSE; character for SE. Name of column in data to use for color basis. Even if fill is the actual ggplot2 aesthetic, the internal code will 'coerce' this column to fill. In order to simplify internal code, MUST be specified.
num_top	numeric. Number of words to show. If between 0 and 1, then assumed to be a percentage.
scale_manual_base	list. Parameters to pass to ggplot2::scale_fill_manual() or ggplot2::scale_color_manual(), depending on the aesthetic mapping. A default values argument is specified in the internal function default_scale_manual(), so if overwriting explicitly, then values should be included in the list.
scale_manual_params	list. Additional parameters to pass to ggplot2::scale_fill_manual() or ggplot2::scale_color_manual() (e.g. breaks, labels, or name). The default values specified by scale_manual_base = default_scale_manual() can be overwritten a values argument in this list (or directly with a values argument in the scale_manual_base list.
labs_base	ggplot2::labs() function. Defaults to a pre-determined set of values. It is recommended NOT to modify this argument; isntead, the _params argument should be used for customization.
labs_params	list. Additional parameters to pass to ggplot2::labs() to use in addition to and/or override the parameters set in the _base parameter.
theme_base	ggplot2::theme() function. (e.g. as ggplot2::theme_minimal().) A custom theme is supplied as a default. As with the the labs_base argument, it is NOT recommended to change this directly.
theme_params	list. Additional parameters to pass to ggplot2::labs(). Should be used in the same manner as labs_params for customization.
facet_base	ggplot2::facet_wrap() function. An internal function sets defaults for facets, scales, ncol, nrow, and strip.position As with the labs_base and theme_base arguments, it is NOT recommended to set this argument explicitly; instead, the accompanying _params argument should be used.
facet_params	list. Additional parameters to pass to ggplot2::facet_wrap(). Parameters passed to this argument override thos in the _base argument.

Value

gg.

Details

Calls compute_tfidf_at() internally. Cannot pass dots to compute_tfidf_at() internally because parameters are used in subsequent processing.

See also

https://www.tidytextmining.com/tfidf.html. https://juliasilge.com/blog/sherlock-holmes-stm/