R: tidy text dataset - tibble

From OnnoWiki
Jump to navigation Jump to search

Text Vector

text <- c("Because I could not stop for Death -",
          "He kindly stopped for me -",
          "The Carriage held but just Ourselves -",
          "and Immortality")
text


Tidy Text Dataset

install.packages("dplyr")
library(dplyr)
text_df <- data_frame(line = 1:4, text = text)
text_df


Tidy Text Novel

library(janeaustenr)
library(dplyr)
library(stringr)
original_books <- austen_books() %>%
   group_by(book) %>%
   mutate(linenumber = row_number(),
          chapter = cumsum(str_detect(text, regex("^chapter [\\divxlc]",
                                            ignore_case = TRUE)))) %>%
   ungroup()
original_books

Buat menjadi one-token-per-row

library(tidytext)
tidy_books <- original_books %>%
              unnest_tokens(word, text)
tidy_books

Buang stopwords

data(stop_words)
tidy_books <- tidy_books %>%
              anti_join(stop_words)

Word Count

tidy_books %>%
    count(word, sort = TRUE)

Plot

library(ggplot2)
tidy_books %>%
    count(word, sort = TRUE) %>%
filter(n > 600) %>%
mutate(word = reorder(word, n)) %>%
ggplot(aes(word, n)) +
       geom_col() +
       xlab(NULL) +
       coord_flip()


Tidy Text gutenbergr

install.packages("gutenbergr")
library(gutenbergr)


hgwells <- gutenberg_download(c(35, 36, 5230, 159))
tidy_hgwells <- hgwells %>%
                unnest_tokens(word, text) %>%
anti_join(stop_words)
tidy_hgwells %>%
    count(word, sort = TRUE)


bronte <- gutenberg_download(c(1260, 768, 969, 9182, 767))
tidy_bronte <- bronte %>%
               unnest_tokens(word, text) %>%
anti_join(stop_words)
tidy_bronte %>%
    count(word, sort = TRUE)

Compare 3 text

library(tidyr)
frequency <- bind_rows(mutate(tidy_bronte, author = "Brontë Sisters"),
                       mutate(tidy_hgwells, author = "H.G. Wells"),
                       mutate(tidy_books, author = "Jane Austen")) %>%
   mutate(word = str_extract(word, "[a-z']+")) %>%
   count(author, word) %>%
   group_by(author) %>%
   mutate(proportion = n / sum(n)) %>%
   select(-n) %>%
   spread(author, proportion) %>%
   gather(author, proportion, `Brontë Sisters`:`H.G. Wells`)

Pranala Menarik