The one with the most words

the-one-with-the-most-words.R

# Load packages ----------------------------------------------------------------

library(tidytext)
library(friends)
library(tidyverse)

# The six friends --------------------------------------------------------------

six_friends <- c("Phoebe Buffay", "Chandler Bing", "Rachel Green", "Monica Geller", "Joey Tribbiani", "Ross Geller")

# Count episodes per season ----------------------------------------------------

episode_count <- friends_info %>%
  count(season, name = "episode_n")

# Make df to plot --------------------------------------------------------------

df <- friends %>%
  filter(speaker != "Scene Directions", speaker != "#ALL#") %>%
  mutate(
    speaker = fct_other(speaker, keep = six_friends, other_level = "Others"),
    speaker = case_when(
      speaker == "Chandler Bing" ~ "Chandler",
      speaker == "Phoebe Buffay" ~ "Phoebe",
      speaker == "Rachel Green" ~ "Rachel",
      speaker == "Monica Geller" ~ "Monica",
      speaker == "Joey Tribbiani" ~ "Joey",
      speaker == "Ross Geller" ~ "Ross",
      TRUE ~ "Others"
    ),
    speaker = fct_relevel(speaker, "Ross", "Rachel", "Phoebe", "Monica", "Joey", "Chandler", "Others")
  ) %>%
  unnest_tokens(word, text) %>%
  # anti_join(stop_words) %>%
  count(season, speaker, sort = TRUE) %>%
  left_join(episode_count) %>%
  mutate(n_avg = n / episode_n)

# Plot -------------------------------------------------------------------------

ggplot(df, aes(x = season, y = n_avg, group = speaker, color = speaker)) +
  geom_smooth(se = FALSE, span = 0.4, show.legend = FALSE) +
  geom_point(size = 0) +
  scale_x_continuous(breaks = 1:10, minor_breaks = NULL) +
  scale_y_continuous(minor_breaks = NULL) +
  scale_color_manual(values = c("#42a2db", "#00944B", "#ff4238", "#9C8CD4", "#ffdc00", "#9a0006", "grey")) +
  theme_minimal() +
  guides(
    color = guide_legend(override.aes = list(size = 5))
  ) +
  theme(
    plot.title = element_text(size = 14, family = "Gabriel Weiss' Friends Font", hjust = 0.5),
    plot.subtitle = element_text(size = 10, family = "Gabriel Weiss' Friends Font"),
    axis.title = element_text(size = 9, family = "Gabriel Weiss' Friends Font"),
    axis.text = element_text(size = 9, family = "Gabriel Weiss' Friends Font"),
    legend.text = element_text(size = 10, family = "Gabriel Weiss' Friends Font"),
    legend.title = element_blank(),
    legend.box.margin = margin(t = 0, r = 0, b = 0, l = 0, unit = "pt"),
    panel.grid.major = element_line(size = 0.3),
    legend.position = "bottom"
  ) +
  labs(
    x = "Season",
    y = "Number of words per episode",
    title = "The one with the most words",
    subtitle = ""
  )

ggsave(filename = "friends.png", width = 7, height = 7 * 0.618, bg = "transparent")