library(dplyr)
library(tidyr)
library(purrr)
library(readr)
library(gutenbergr)
library(tidytext)
library(ggplot2)
#a---------
komen1 <- read_file(file.choose()) #pilih komen1
komen2 <- read_file(file.choose()) #pilih komen2
komen3 <- read_file(file.choose()) #pilih komen3
komen4 <- read_file(file.choose()) #pilih komen4

#b------
komen1_df <- tibble(line=1:1,text = komen1)
komen1_text <- komen1_df %>% unnest_tokens(word, text)

komen2_df <- tibble(line=1:1,text = komen2)
komen2_text <- komen2_df %>% unnest_tokens(word, text)

komen3_df <- tibble(line=1:1,text = komen3)
komen3_text <- komen3_df %>% unnest_tokens(word, text)

komen4_df <- tibble(line=1:1,text = komen4)
komen4_text <- komen4_df %>% unnest_tokens(word, text)
#c--------
data(stop_words)

komen1_clean <- komen1_text %>%
 anti_join(stop_words,by="word")

komen1_clean %>%
 count(word, sort = TRUE) %>%
 filter(n > 1) %>%
 mutate(word = reorder(word, n)) %>%
 ggplot(aes(word, n)) +
 geom_col() +
 xlab(NULL) +
 coord_flip()

komen2_clean <- komen2_text %>%
 anti_join(stop_words,by="word")
komen2_clean %>%
 count(word, sort = TRUE) %>%
 filter(n > 1) %>%
 mutate(word = reorder(word, n)) %>%
 ggplot(aes(word, n)) +
 geom_col() +
 xlab(NULL) +
 coord_flip()

komen3_clean <- komen3_text %>%
 anti_join(stop_words,by="word")
komen3_clean %>%
 count(word, sort = TRUE) %>%
 filter(n > 1) %>%
 mutate(word = reorder(word, n)) %>%
 ggplot(aes(word, n)) +
 geom_col() +
 xlab(NULL) +
 coord_flip()

komen4_clean <- komen4_text %>%
 anti_join(stop_words,by="word")
komen4_clean %>%
 count(word, sort = TRUE) %>%
 filter(n > 1) %>%
 mutate(word = reorder(word, n)) %>%
 ggplot(aes(word, n)) +
 geom_col() +
 xlab(NULL) +
 coord_flip()


#d-----
#Komen1
komen1_sentiment <- komen1_clean %>%
 inner_join(get_sentiments("bing"),by="word") %>%
 count(word, index = line, sentiment) %>%
 spread(sentiment, n, fill = 0) %>%
   mutate(sentiment = positive - negative)

komen1_sentiment

komen1_counts <- komen1_clean %>%
  inner_join(get_sentiments("bing")) %>%
  count(word, sentiment, sort = TRUE) %>%
  ungroup()

komen1_counts

komen1_counts %>%
  group_by(sentiment) %>%
  top_n(3) %>%
  ungroup() %>%
  mutate(word = reorder(word, n)) %>%
  ggplot(aes(word, n, fill = sentiment)) +
  geom_col(show.legend = FALSE) +
  facet_wrap(~sentiment, scales = "free_y") +
  labs(y = "Contribution to sentiment",
       x = NULL) +
  coord_flip()

#Komen2
komen2_sentiment <- komen2_clean %>%
 inner_join(get_sentiments("bing"),by="word") %>%
 count(word, index = line, sentiment) %>%
 spread(sentiment, n, fill = 0) %>%
   mutate(sentiment = positive - negative)

komen2_sentiment

komen2_counts <- komen2_clean %>%
  inner_join(get_sentiments("bing")) %>%
  count(word, sentiment, sort = TRUE) %>%
  ungroup()

komen2_counts

komen2_counts %>%
  group_by(sentiment) %>%
  top_n(5) %>%
  ungroup() %>%
  mutate(word = reorder(word, n)) %>%
  ggplot(aes(word, n, fill = sentiment)) +
  geom_col(show.legend = FALSE) +
  facet_wrap(~sentiment, scales = "free_y") +
  labs(y = "Contribution to sentiment",
       x = NULL) +
  coord_flip()

#Komen3
komen3_sentiment <- komen3_clean %>%
 inner_join(get_sentiments("bing"),by="word") %>%
 count(word, index = line, sentiment) %>%
 spread(sentiment, n, fill = 0) %>%
   mutate(sentiment = positive - negative)

komen3_sentiment

komen3_counts <- komen3_clean %>%
  inner_join(get_sentiments("bing")) %>%
  count(word, sentiment, sort = TRUE) %>%
  ungroup()

komen3_counts

komen3_counts %>%
  group_by(sentiment) %>%
  top_n(4) %>%
  ungroup() %>%
  mutate(word = reorder(word, n)) %>%
  ggplot(aes(word, n, fill = sentiment)) +
  geom_col(show.legend = FALSE) +
  facet_wrap(~sentiment, scales = "free_y") +
  labs(y = "Contribution to sentiment",
       x = NULL) +
  coord_flip()

#Komen4
komen4_sentiment <- komen4_clean %>%
 inner_join(get_sentiments("bing"),by="word") %>%
 count(word, index = line, sentiment) %>%
 spread(sentiment, n, fill = 0) %>%
   mutate(sentiment = positive - negative)

komen4_sentiment

komen4_counts <- komen4_clean %>%
  inner_join(get_sentiments("bing")) %>%
  count(word, sentiment, sort = TRUE) %>%
  ungroup()

komen4_counts

komen4_counts %>%
  group_by(sentiment) %>%
  top_n(5) %>%
  ungroup() %>%
  mutate(word = reorder(word, n)) %>%
  ggplot(aes(word, n, fill = sentiment)) +
  geom_col(show.legend = FALSE) +
  facet_wrap(~sentiment, scales = "free_y") +
  labs(y = "Contribution to sentiment",
       x = NULL) +
  coord_flip()