word cloud

R code of the ggplot2 example

library(ggplot2)
library(dplyr)
library(tidytext)
library(ggwordcloud)
library(HistData)

## The 60 most frequent words of the 85 essays, common English words left out.
words <- Federalist_text |>
  unnest_tokens(word, text) |>
  anti_join(stop_words, by = "word") |>
  count(word, sort = TRUE) |>
  slice_head(n = 60) |>
  mutate(top = row_number() == 1)

set.seed(1)
p <- ggplot(words, aes(label = word, size = n, colour = top)) +
  geom_text_wordcloud(family = "Quicksand", eccentricity = 0.6, rm_outside = TRUE) +
  scale_size_area(max_size = 9) +
  scale_colour_manual(values = c(`FALSE` = "#535353", `TRUE` = "#9b2367")) +
  labs(title = "“Government” is the most frequent word of The Federalist",
       subtitle = "60 most frequent words of the 85 essays, 1787–1788, size by count",
       caption = "Data: The Federalist Papers, via the HistData R package") +
  theme_lex_viz(grid = "", arrows = "") +
  theme(axis.text = element_blank())