Skip to contents

Visualize output

Source: R/visualize.R
visualize.Rd

visualize() provides a simple method for displaying results. Based on previous functions used, visualize() will choose a method, resolving to one of the other visualizing helpers.

Usage

visualize(.data, ...)

Arguments

.data

data processed with one or more functions from tmtyro

...

Arguments passed on to plot_doc_word_bars, plot_bigrams, plot_vocabulary, plot_ttr, plot_hir, plot_topic_distributions, plot_topic_bars, plot_topic_wordcloud

rows

The features to show

by

The column used for document grouping, with doc_id as the default

feature

The column to measure, as in "word" or "lemma"

inorder

Whether to retain the factor order of the "by" column

reorder_y

Whether to reorder the Y-values by facet

color_y

Whether bars should be filled by Y-values

percents

Whether to display word frequencies as percentage instead of raw counts

label

Whether to show the value as a label with each bar

label_tweak

The numeric value by which to tweak the label, if shown. For percentages, this value adjusts the decimal-point precision. For raw counts, this value adjusts labels' offset from the bars

label_inside

Whether to show the value as a label inside each bar

na_rm

Whether to drop empty features

random_seed

Whether to randomize the creation of the network chart.

set_seed

A specific seed to use if not random

legend

Whether to show a legend for the edge color

top_n

The number of pairs to visualize

identity

A grouping column for lines

descriptive_labels

A toggle for disabling descriptive labels of progress_percent on the X-axis

labeling

Options for labeling groups:

  • "point" labels the final value

  • "inline" prints the label within a smoothed curve

  • "axis" prints labels where a secondary Y-axis might go

  • "inset" prints a legend within the plot area

  • Anything else prints a legend to the right of the plot area.

log_y

A toggle for logarithmic scaling to the Y-axis; defaults to TRUE

topics

The topic numbers to view

Value

a ggplot2 object

Note

For some visualizations, an optional type parameter may be helpful to change the visualization. For example, setting type = "htr", type = "ttr", or type = "hapax" after add_vocabulary() will emphasize different columns added by that function. Similarly, type = "cloud" or type = "wordcloud" will show topic word clouds after make_topic_model(), and type = "heatmap" will show an alternative visualization for word frequencies.

See also

Other visualizing helpers: change_colors(), plot_bigrams(), plot_doc_word_bars(), plot_doc_word_heatmap(), plot_hapax(), plot_hir(), plot_tf_idf(), plot_topic_bars(), plot_topic_distributions(), plot_topic_wordcloud(), plot_ttr(), plot_vocabulary()

Examples

dubliners <- get_gutenberg_corpus(2814) |>
  load_texts() |>
  identify_by(part) |>
  standardize_titles()

# A data frame with `doc_id` and `word` columns will visualize frequency by default
dubliners |>
   visualize()


# Applying `tmtyro` functions will choose an appropriate visualization

dubliners |>
   add_ngrams() |>
   visualize()


dubliners |>
   add_ngrams() |>
   combine_ngrams() |>
   visualize()


dubliners |>
   summarize_tf_idf() |>
   visualize()


dubliners |>
   add_vocabulary() |>
   visualize()


if (FALSE) { # sentiment requires interaction on first load
  dubliners |>
     add_sentiment() |>
     visualize()
}

# Some visualizations are specified with the `type` argument
dubliners |>
   add_vocabulary() |>
   visualize(type = "ttr")


if (FALSE) { # puzzlingly broken for Dubliners, but usually works
dubliners |>
   add_vocabulary() |>
   visualize(type = "hapax")
}

# Other arguments get passed along
dubliners |>
   add_ngrams() |>
   visualize(top_n = 25)


dubliners |>
   add_vocabulary() |>
   visualize(x = progress_percent)