War Letters

install.packages(c("tidyverse","tidytext","pdftools","tesseract","tidyr","stringr","ggplot2"))
install.packages("tidyr")
install.packages("devtools")
install.packages("textdata")

library(tidyverse)
library(pdftools)
library(tesseract)
library(tidyr)
library(devtools)
library(textdata)
library(tidytext)
library(dplyr)
library(stringr)
library(ggplot2)

install.packages("ggpubr")
library(ggpubr)

txt = 0
txt <- read_file(paste("C:\\Users\\juand\\Documents\\RLAC ASS 2\\", "americanwoman.txt", sep =""))

txt_sentences <- tibble(text = txt) %>% 
  unnest_tokens(sentence, text, token = "sentences") 

d = nrow(txt_sentences)

tidy_text <- txt_sentences %>% 
  add_column(linenumber = 1:d, .before = "sentence")

tidy_text <- tidy_text %>%
  unnest_tokens(word, sentence) %>%
  anti_join(stop_words)

Sentiment analysis bit

text_sentiment = 0
a = 50

text_sentiment <- tidy_text %>%
  inner_join(get_sentiments("bing"))

Error in tidy_text %>% inner_join(get_sentiments("bing")) : 
  could not find function "%>%"

Now that we know that this works, we do it for the five books

  books <- c('americanwoman.txt','carryon.txt','lettersfromfrance.txt','lettersfromschoolboy.txt','livingdeadman.txt')

plots <- list()

for(book in books)
{
  txt <- read_file(paste("C:\\Users\\juand\\Documents\\RLAC ASS 2\\", book, sep =""))
  txt_sentences <- tibble(text = txt) %>% 
  unnest_tokens(sentence, text, token = "sentences") 

d = nrow(txt_sentences)

tidy_text <- txt_sentences %>% 
  add_column(linenumber = 1:d, .before = "sentence")

tidy_text <- tidy_text %>%
  unnest_tokens(word, sentence) %>%
  anti_join(stop_words)
text_sentiment = 0
a = 50

text_sentiment <- tidy_text %>%
  inner_join(get_sentiments("bing"))

text_sentiment$linenumber <- text_sentiment$linenumber%/%a

text_sentiment <- text_sentiment %>% count(index= linenumber,sentiment) %>% spread(key = sentiment, value = n)

text_sentiment$total = NA

text_sentiment$total <- text_sentiment$positive - text_sentiment$negative

P <- ggplot(text_sentiment, aes(index, total, colour = total)) +
   geom_point(position=position_jitter(0.3)) + geom_smooth() + xlab(book)

plots[[length(plots) + 1]] <- P
  
}

Joining, by = "word"
Joining, by = "word"
Joining, by = "word"
Joining, by = "word"
Joining, by = "word"
Joining, by = "word"
Joining, by = "word"
Joining, by = "word"
Joining, by = "word"
Joining, by = "word"

ggarrange(plotlist = plots, ncol = 2)

`geom_smooth()` using method = 'loess' and formula 'y ~ x'
`geom_smooth()` using method = 'loess' and formula 'y ~ x'
`geom_smooth()` using method = 'loess' and formula 'y ~ x'
`geom_smooth()` using method = 'loess' and formula 'y ~ x'
`geom_smooth()` using method = 'loess' and formula 'y ~ x'

$`1`

$`2`

$`3`

attr(,"class")
[1] "list"      "ggarrange"

OK ALL THE LETTERS SEEM TO GO DOWN AT THE 25 to 50% mark of the book. We can take a sample of that and one of the rest of the books and see what the differences areeeee