Some tidytext anaylysis of sea chanties

chanty_analysis.R

library(tidyverse)
library(tidytext)
library(rvest)
library(ggplot)

#get the initial info from the site
site <- "http://www.guscantieni.com/bindarrr/"

index <- read_html(paste0(site, "index.html"))

links <- html_nodes(index, "a") %>%
  html_attr("href") %>%
  keep(~str_detect(.x, "^songs")) %>%
  discard(~str_detect(.x, "table-of-contents"))  %>%
  discard(~str_detect(.x, "songs/what-will-we-do\\?.html"))  #broken
  

#function to get the text
keelhaul_one_page <- function(a_link){
  print(a_link)
  a_chanty <- read_html(paste0(site, a_link))
  title <-  verses <- a_chanty %>% html_nodes("h1") %>%
    html_text
  
  verses <- a_chanty %>% html_nodes("p") %>%
    html_text
  
  if(length(verses)==0)
    verses <- a_chanty %>% html_nodes("td") %>%
    html_text
  
  tibble(chanty = title, verse = 1:length(verses), words = verses)
}

#get the lyrics and tokenize
chanties <- map_df(links, keelhaul_one_page) 

chanty_words <- chanties %>%
  unnest_tokens(word, words) %>%
  anti_join(get_stopwords())

#count words and ascribe sentiments across all
bing <- get_sentiments("bing")

wordcount <- chanty_words %>%
  count(word, sort=TRUE) %>%
  inner_join(bing) 


ggplot(wordcount %>%  
         filter(n>10) %>%
         mutate(n = ifelse(sentiment == "negative", -n, n))  %>%
         mutate(word = reorder(word, n)),
       aes (x = word, y = n, fill = sentiment)) +
  geom_col() +
  coord_flip() +
  ggtitle("Word contribution to sentiment of chanties\nlisted on http://www.guscantieni.com/bindarrr/")