library(tidytext)Warning: package 'tidytext' was built under R version 4.5.3library(janeaustenr)Warning: package 'janeaustenr' was built under R version 4.5.3library(dplyr)
Attaching package: 'dplyr'The following objects are masked from 'package:stats':
filter, lagThe following objects are masked from 'package:base':
intersect, setdiff, setequal, unionlibrary(stringr)
library(tidyr)
library(ggplot2)tidy_books <- austen_books() %>%
group_by(book) %>%
mutate(
linenumber = row_number(),
chapter = cumsum(str_detect(
text,
regex("^chapter [\\divxlc]", ignore_case = TRUE)
))
) %>%
ungroup() %>%
unnest_tokens(word, text)
jane_austen_sentiment <- tidy_books %>%
inner_join(get_sentiments("bing")) %>%
count(book, index = linenumber %/% 80, sentiment) %>%
pivot_wider(names_from = sentiment, values_from = n, values_fill = 0) %>%
mutate(sentiment = positive - negative)Joining with `by = join_by(word)`Warning in inner_join(., get_sentiments("bing")): Detected an unexpected many-to-many relationship between `x` and `y`.
ℹ Row 435434 of `x` matches multiple rows in `y`.
ℹ Row 5051 of `y` matches multiple rows in `x`.
ℹ If a many-to-many relationship is expected, set `relationship =
"many-to-many"` to silence this warning.ggplot(jane_austen_sentiment, aes(index, sentiment, fill = book)) +
geom_col(show.legend = FALSE) +
facet_wrap(~book, ncol = 2, scales = "free_x")
I’ll use AFINN instead of BING. I’ll also try the gutenbergr package
library(gutenbergr)Warning: package 'gutenbergr' was built under R version 4.5.3gutenberg_metadata |> filter(str_detect(author, "Vonnegut"))# A tibble: 2 × 8
gutenberg_id title author gutenberg_author_id language gutenberg_bookshelf
<int> <chr> <chr> <int> <fct> <chr>
1 21279 2 B R 0 … Vonne… 9812 en Science Fiction/Ca…
2 30240 The Big … Vonne… 9812 en Science Fiction/Ca…
# ℹ 2 more variables: rights <fct>, has_text <lgl>vonnegut <- gutenberg_download(30240)Using mirror https://aleph.pglaf.org.head(vonnegut)# A tibble: 6 × 2
gutenberg_id text
<int> <chr>
1 30240 " THE BIG TRIP"
2 30240 " UP YONDER"
3 30240 ""
4 30240 "By KURT VONNEGUT, JR."
5 30240 ""
6 30240 "Illustrated by KOSSIN"I’ll also use AFINN
afinn <- get_sentiments("afinn")# transform Vonnegut file to get each line, chapter, word as it's own row.
df_vonnegut <- vonnegut |>
mutate(
linenumber = row_number(),
chapter = cumsum(str_detect(
text,
regex("^chapter [\\divxlc]", ignore_case = TRUE)
))
) |>
ungroup() |>
unnest_tokens(word, text)
# inner join the afinn data values per word
vonnegut_sentiment <- df_vonnegut |>
inner_join(afinn, by = "word") |>
mutate(index = linenumber %/% 80) |>
group_by(index) |>
summarise(
sentiment = sum(value, na.rm = TRUE), # net AFINN score
n_words = n() # words with sentiment scores
) |>
ungroup()
vonnegut_sentiment |>
ggplot(aes(x = index, y = sentiment)) +
geom_col(fill = "darkred", alpha = 0.8) +
geom_smooth(se = FALSE, color = "black") +
labs(title = "Sentiment Progression in the Vonnegut Book (AFINN)",
x = "Section (every 80 lines)",
y = "Net Sentiment Score") +
theme_minimal()`geom_smooth()` using method = 'loess' and formula = 'y ~ x'
This is a tiny book / short story from Vonnegut that is dark and satirical. Using AFINN, we see the darker sentiment in the middle when the family conflict gets worse and everything becomes more hostile and tense. Then we see more levity near the end, when Lou and Em decide they actually like being in prison because it finally gives them peace and privacy. That shift fits the story well, so it is interesting that AFINN is able to follow the basic emotional arc even if it cannot fully understand the satire.