mueller-report_top-25-words.R

library(tidyverse)
library(tidytext)

mueller_report <- read_csv("mueller_report.csv")

mueller_report_tidy <- 
  mueller_report %>% 
  tidytext::unnest_tokens(word, text)

mueller_report_tidy %>% 
  filter(line > 2) %>% 
  anti_join(tidytext::stop_words) %>% 
  filter(!is.na(word), str_detect(word, "[a-zA-Z]")) %>% 
  mutate(word = str_remove(word, "'s$")) %>% 
  mutate(word = str_replace(word, "^corney$", "comey")) %>% 
  mutate(word = str_replace(word, "russian?s?", "russia")) %>% 
  group_by(word) %>% 
  count(word, sort = TRUE) %>%
  ungroup() %>% 
  mutate(word = fct_rev(fct_inorder(word))) %>% 
  slice(1:25) %>% 
  ggplot() +
  aes(word, n, fill = n) +
  geom_col() +
  coord_flip() +
  labs(y = "Number of Times Mentioned", x = NULL, 
       title = "25 Most-Used Words in the Redacted Mueller Report",
       caption = "Chart:   @grrrck\nData:    github.com/gadenbuie/mueller-report\nSource: Steven Rich (Washington Post)") +
  scale_y_continuous(expand = c(0.01,0), limits = c(0, 2500)) +
  scale_fill_gradient(low = "#b7c6d6", high = "#445566", guide = FALSE) +
  # ggsci::scale_fill_material("deep-purple") +
  grkmisc::theme_grk(panel_background_color = NA) +
  theme(
    panel.border = element_blank(),
    panel.background = element_blank(),
    panel.grid.major.y = element_blank(),
    plot.caption = element_text(hjust = 0)
  )