Difference between revisions of "R: tidytext: tf-idf corpus of physics texts"

From OnnoWiki
Jump to: navigation, search
(Created page with "# Ref: https://github.com/dgrtwo/tidy-text-mining/blob/master/03-tf-idf.Rmd library(knitr) opts_chunk$set(message = FALSE, warning = FALSE, cache = TRUE) options(width =...")
 
 
Line 1: Line 1:
# Ref: https://github.com/dgrtwo/tidy-text-mining/blob/master/03-tf-idf.Rmd
+
# Ref: https://github.com/dgrtwo/tidy-text-mining/blob/master/03-tf-idf.Rmd
  
  

Latest revision as of 04:48, 4 December 2019

# Ref: https://github.com/dgrtwo/tidy-text-mining/blob/master/03-tf-idf.Rmd


library(knitr)
opts_chunk$set(message = FALSE, warning = FALSE, cache = TRUE)
options(width = 100, dplyr.width = 100)
library(ggplot2)
theme_set(theme_light())


# gutenbergr
library(gutenbergr)
physics <- gutenberg_download(c(37729, 14725, 13476, 30155), 
                              meta_fields = "author")
# count
physics_words <- physics %>%
  unnest_tokens(word, text) %>%
  count(author, word, sort = TRUE)
physics_words


# calculate & plot
library(forcats)
plot_physics <- physics_words %>%
  bind_tf_idf(word, author, n) %>%
  mutate(word = fct_reorder(word, tf_idf)) %>%
  mutate(author = factor(author, levels = c("Galilei, Galileo",
                                            "Huygens, Christiaan", 
                                            "Tesla, Nikola",
                                            "Einstein, Albert")))
plot_physics %>% 
  group_by(author) %>% 
  top_n(15, tf_idf) %>% 
  ungroup() %>%
  mutate(word = reorder(word, tf_idf)) %>%
  ggplot(aes(word, tf_idf, fill = author)) +
  geom_col(show.legend = FALSE) +
  labs(x = NULL, y = "tf-idf") +
  facet_wrap(~author, ncol = 2, scales = "free") +
  coord_flip()


# evaluasi _k_
library(stringr)
physics %>% 
  filter(str_detect(text, "_k_")) %>% 
  select(text)


# clean up
physics %>% 
  filter(str_detect(text, "RC")) %>% 
  select(text)


# stopwords & plot
mystopwords <- tibble(word = c("eq", "co", "rc", "ac", "ak", "bn", 
                               "fig", "file", "cg", "cb", "cm",
                               "ab", "_k", "_k_", "_x"))
physics_words <- anti_join(physics_words, mystopwords, 
                           by = "word")
plot_physics <- physics_words %>%
  bind_tf_idf(word, author, n) %>%
  mutate(word = str_remove_all(word, "_")) %>%
  group_by(author) %>% 
  top_n(15, tf_idf) %>%
  ungroup() %>%
  mutate(word = reorder_within(word, tf_idf, author)) %>%
  mutate(author = factor(author, levels = c("Galilei, Galileo",
                                            "Huygens, Christiaan",
                                            "Tesla, Nikola",
                                            "Einstein, Albert")))
ggplot(plot_physics, aes(word, tf_idf, fill = author)) +
  geom_col(show.legend = FALSE) +
  labs(x = NULL, y = "tf-idf") +
  facet_wrap(~author, ncol = 2, scales = "free") +
  coord_flip() +
  scale_x_reordered()


Referensi

Pranala Menarik