Text and Sentiment Analysis in R

Sentiment Analysis

Exercise 1

library(tidyverse)
library(tidytext)
library(gutenbergr)

data(stop_words)

hound <- gutenberg_download(2852) %>% 
           unnest_tokens(word, text) %>% 
           anti_join(stop_words)

sentiments <- get_sentiments("nrc")

hound_sentiments <- hound %>% inner_join(sentiments)
                      
hound_sentiments %>%
  count(sentiment) %>%
  mutate(sentiment = reorder(sentiment, n)) %>%
  ggplot(aes(n, sentiment)) + geom_col() + labs(y=NULL)

Exercise 2

hound <- gutenberg_download(2852) %>% 
           mutate(linenumber=row_number()) %>% 
           unnest_tokens(word, text) %>% 
           anti_join(stop_words)

hound_sentiments <- hound %>% 
                      inner_join(sentiments) %>% 
                      count(block=linenumber%/%80, sentiment)

hound_sentiments <- hound_sentiments %>% 
                      pivot_wider(names_from = sentiment, 
                                  values_from = n, 
                                  values_fill = 0)

hound_sentiments <- hound_sentiments %>%
                      mutate(sentiment = positive - negative)

hound_sentiments %>% ggplot(aes(block, sentiment)) + geom_col()

Exercise 3

block_size = 50

hound %>% 
  inner_join(sentiments) %>% 
  count(block=linenumber%/%block_size, sentiment) %>%
  pivot_wider(names_from = sentiment, 
              values_from = n, 
              values_fill = 0) %>%
  mutate(sentiment = positive - negative) %>%
  ggplot(aes(block, sentiment)) + geom_col()

Return