R: bigram - Revision history

Onnowpurbo at 00:46, 5 November 2018

2018-11-05T00:46:54Z

← Older revision		Revision as of 00:46, 5 November 2018
Line 15:		Line 15:
	bigrams_separated <- text_bigrams %>%		bigrams_separated <- text_bigrams %>%
	separate(bigram, c("word1", "word2"), sep = " ")		separate(bigram, c("word1", "word2"), sep = " ")
			#
	# stopwords default		# stopwords default
	bigrams_filtered <- bigrams_separated %>%		bigrams_filtered <- bigrams_separated %>%
	filter(!word1 %in% stop_words$word) %>%		filter(!word1 %in% stop_words$word) %>%
	filter(!word2 %in% stop_words$word)		filter(!word2 %in% stop_words$word)
			#
	# stopwords Indonesia		# stopwords Indonesia
	bigrams_filtered <- bigrams_separated %>%		bigrams_filtered <- bigrams_separated %>%
Line 25:		Line 27:
	bigram_counts <- bigrams_filtered %>%		bigram_counts <- bigrams_filtered %>%
	count(word1, word2, sort = TRUE)		count(word1, word2, sort = TRUE)
			#
	bigram_graph <- bigram_counts %>%		bigram_graph <- bigram_counts %>%
	filter(n > 40) %>%		filter(n > 40) %>%
Line 34:		Line 37:
	geom_node_point() +		geom_node_point() +
	geom_node_text(aes(label = name), vjust = 1, hjust = 1)		geom_node_text(aes(label = name), vjust = 1, hjust = 1)
			#

	bigrams_united <- bigrams_filtered %>%		bigrams_united <- bigrams_filtered %>%
	unite(bigram, word1, word2, sep = " ")		unite(bigram, word1, word2, sep = " ")
	bigrams_united		bigrams_united
			#
	bigram_tf_idf <- bigrams_united %>%		bigram_tf_idf <- bigrams_united %>%
	count(doc_id, bigram) %>%		count(doc_id, bigram) %>%

2018-11-05T00:29:04Z

← Older revision		Revision as of 00:29, 5 November 2018
Line 41:		Line 41:

	bigram_tf_idf <- bigrams_united %>%		bigram_tf_idf <- bigrams_united %>%
	count(~~book~~, bigram) %>%		count(doc_id, bigram) %>%
	bind_tf_idf(~~book~~, bigram, n) %>%		bind_tf_idf(doc_id, bigram, n) %>%
	arrange(desc(tf_idf))		arrange(desc(tf_idf))
	bigram_tf_idf		bigram_tf_idf
Line 66:		Line 66:
	bigram_counts <- bigrams_filtered %>%		bigram_counts <- bigrams_filtered %>%
	count(word1, word2, sort = TRUE)		count(word1, word2, sort = TRUE)




	bigrams_united <- bigrams_filtered %>%		bigrams_united <- bigrams_filtered %>%

2018-11-04T13:48:26Z

← Older revision		Revision as of 13:48, 4 November 2018
Line 35:		Line 35:
	geom_node_text(aes(label = name), vjust = 1, hjust = 1)		geom_node_text(aes(label = name), vjust = 1, hjust = 1)


			bigrams_united <- bigrams_filtered %>%
			unite(bigram, word1, word2, sep = " ")
			bigrams_united

			bigram_tf_idf <- bigrams_united %>%
			count(book, bigram) %>%
			bind_tf_idf(book, bigram, n) %>%
			arrange(desc(tf_idf))
			bigram_tf_idf

2018-11-04T13:43:54Z

2018-11-04T13:19:07Z

← Older revision		Revision as of 13:19, 4 November 2018
Line 14:		Line 14:
	bigrams_separated <- text_bigrams %>%		bigrams_separated <- text_bigrams %>%
	separate(bigram, c("word1", "word2"), sep = " ")		separate(bigram, c("word1", "word2"), sep = " ")
			# stopwords default
	bigrams_filtered <- bigrams_separated %>%		bigrams_filtered <- bigrams_separated %>%
	filter(!word1 %in% stop_words$word) %>%		filter(!word1 %in% stop_words$word) %>%
	filter(!word2 %in% stop_words$word)		filter(!word2 %in% stop_words$word)
			# stopwords Indonesia
			bigrams_filtered <- bigrams_separated %>%
			filter(!word1 %in% stopwords::stopwords("id", source = "stopwords-iso")) %>%
			filter(!word2 %in% stopwords::stopwords("id", source = "stopwords-iso"))
	bigram_counts <- bigrams_filtered %>%		bigram_counts <- bigrams_filtered %>%
	count(word1, word2, sort = TRUE)		count(word1, word2, sort = TRUE)

2018-11-04T13:13:30Z

2018-11-04T13:13:14Z

← Older revision		Revision as of 13:13, 4 November 2018
Line 3:		Line 3:
	library(tidytext)		library(tidytext)
	library(janeaustenr)		library(janeaustenr)
			library(tidyr)
			library(igraph)
			library(ggplot2)
			library(ggraph)
			library(ggraph)

	text <- readtext("out.txt")		text <- readtext("out.txt")
Line 9:		Line 13:
	unnest_tokens(bigram, text, token = "ngrams", n = 2)		unnest_tokens(bigram, text, token = "ngrams", n = 2)
	text_bigrams		text_bigrams
	~~library(tidyr)~~
	bigrams_separated <- text_bigrams %>%		bigrams_separated <- text_bigrams %>%
	separate(bigram, c("word1", "word2"), sep = " ")		separate(bigram, c("word1", "word2"), sep = " ")
Line 17:		Line 20:
	bigram_counts <- bigrams_filtered %>%		bigram_counts <- bigrams_filtered %>%
	count(word1, word2, sort = TRUE)		count(word1, word2, sort = TRUE)
	~~library(igraph)~~
	~~library(ggplot2)~~
	~~library(ggraph)~~
	bigram_graph <- bigram_counts %>%		bigram_graph <- bigram_counts %>%
	filter(n > 40) %>%		filter(n > 40) %>%
	graph_from_data_frame()		graph_from_data_frame()
	bigram_graph		bigram_graph
	~~library(ggraph)~~
	set.seed(2017)		set.seed(2017)
	ggraph(bigram_graph, layout = "fr") +		ggraph(bigram_graph, layout = "fr") +

2018-11-04T13:10:36Z

← Older revision		Revision as of 13:10, 4 November 2018
Line 17:		Line 17:
	bigram_counts <- bigrams_filtered %>%		bigram_counts <- bigrams_filtered %>%
	count(word1, word2, sort = TRUE)		count(word1, word2, sort = TRUE)
			library(igraph)
			library(ggplot2)
			library(ggraph)
			bigram_graph <- bigram_counts %>%
			filter(n > 40) %>%
			graph_from_data_frame()
			bigram_graph
			library(ggraph)
			set.seed(2017)
			ggraph(bigram_graph, layout = "fr") +
			geom_edge_link() +
			geom_node_point() +
			geom_node_text(aes(label = name), vjust = 1, hjust = 1)

2018-11-04T12:59:16Z

@@ Line 4: / Line 4: @@
   library(janeaustenr)
   austen_bigrams <- austen_books() %>%
                     unnest_tokens(bigram, text, token = "ngrams", n = 2)
@@ Line 10: / Line 28: @@
   austen_bigrams %>%
         count(bigram, sort = TRUE)
   library(tidyr)

2018-10-31T05:43:00Z