R: tidytext NASA data - Revision history

Onnowpurbo at 05:25, 8 November 2018

2018-11-08T05:25:31Z

← Older revision		Revision as of 05:25, 8 November 2018
Line 3:		Line 3:
	metadata <- fromJSON("https://data.nasa.gov/data.json")		metadata <- fromJSON("https://data.nasa.gov/data.json")
	names(metadata$dataset)		names(metadata$dataset)
			save(metadata,file="metadata.Rdata")


	class(metadata$dataset$title)		class(metadata$dataset$title)
Line 13:		Line 15:
	title = metadata$dataset$title)		title = metadata$dataset$title)
	nasa_title		nasa_title
			save(nasa_title,file="nasa_title.Rdata")



Line 29:		Line 33:
	unnest(keyword)		unnest(keyword)
	nasa_keyword		nasa_keyword
			save(nasa_desc,file="nasa_desc.Rdata")
			save(nasa_keyword,file="nasa_keyword.Rdata")

Onnowpurbo at 03:46, 8 November 2018

2018-11-08T03:46:28Z

← Older revision		Revision as of 03:46, 8 November 2018
Line 70:		Line 70:
	title_word_pairs		title_word_pairs

			# ini bisa gagal karena terlalu besar
			desc_word_pairs <- nasa_desc %>%
			pairwise_count(word, id, sort = TRUE, upper = FALSE)
			desc_word_pairs


			library(ggplot2)
			library(igraph)
			library(ggraph)
			set.seed(1234)
			title_word_pairs %>%
			filter(n >= 250) %>%
			graph_from_data_frame() %>%
			ggraph(layout = "fr") +
			geom_edge_link(aes(edge_alpha = n, edge_width = n), edge_colour = "cyan4") +
			geom_node_point(size = 5) +
			geom_node_text(aes(label = name), repel = TRUE,
			point.padding = unit(0.2, "lines")) +
			theme_void()

Onnowpurbo at 01:45, 8 November 2018

2018-11-08T01:45:29Z

← Older revision		Revision as of 01:45, 8 November 2018
Line 63:		Line 63:
	mutate(keyword = toupper(keyword))		mutate(keyword = toupper(keyword))


			#
			library(widyr)
			title_word_pairs <- nasa_title %>%
			pairwise_count(word, id, sort = TRUE, upper = FALSE)
			title_word_pairs

Onnowpurbo at 01:19, 8 November 2018

2018-11-08T01:19:41Z

← Older revision		Revision as of 01:19, 8 November 2018
Line 29:		Line 29:
	unnest(keyword)		unnest(keyword)
	nasa_keyword		nasa_keyword



Line 43:		Line 42:


			# count & sort
	nasa_title %>%		nasa_title %>%
	count(word, sort = TRUE)		count(word, sort = TRUE)
Line 50:		Line 49:


			# buang keyword yang tidak ada artinya
	my_stopwords <- data_frame(word = c(as.character(1:10),		my_stopwords <- data_frame(word = c(as.character(1:10),
	"v1", "v03", "l2", "l3", "l4", "v5.2.0",		"v1", "v03", "l2", "l3", "l4", "v5.2.0",
Line 59:		Line 59:


			# supaya OCEAN & Ocean jadi sama
			nasa_keyword <- nasa_keyword %>%
			mutate(keyword = toupper(keyword))

Onnowpurbo at 01:17, 8 November 2018

2018-11-08T01:17:10Z

@@ Line 39: / Line 39: @@
       unnest_tokens(word, desc) %>%
          anti_join(stop_words)
   nasa_title
   nasa_desc

Onnowpurbo at 01:13, 8 November 2018

2018-11-08T01:13:32Z

← Older revision		Revision as of 01:13, 8 November 2018
Line 30:		Line 30:
	nasa_keyword		nasa_keyword



			library(tidytext)
			nasa_title <- nasa_title %>%
			unnest_tokens(word, title) %>%
			anti_join(stop_words)
			nasa_desc <- nasa_desc %>%
			unnest_tokens(word, desc) %>%
			anti_join(stop_words)

			nasa_title
			nasa_desc

Onnowpurbo at 01:04, 8 November 2018

2018-11-08T01:04:52Z

← Older revision		Revision as of 01:04, 8 November 2018
Line 22:		Line 22:
	sample_n(5)		sample_n(5)


			# cari keyword
			library(tidyr)
			nasa_keyword <- data_frame(id = metadata$dataset$`_id`$`$oid`,
			keyword = metadata$dataset$keyword) %>%
			unnest(keyword)
			nasa_keyword

Onnowpurbo at 01:03, 8 November 2018

2018-11-08T01:03:48Z

← Older revision		Revision as of 01:03, 8 November 2018
Line 8:		Line 8:
	class(metadata$dataset$keyword)		class(metadata$dataset$keyword)

			# ambil judul2
	library(dplyr)		library(dplyr)
	nasa_title <- data_frame(id = metadata$dataset$`_id`$`$oid`,		nasa_title <- data_frame(id = metadata$dataset$`_id`$`$oid`,
Line 14:		Line 15:


			# sampling 5 line
			nasa_desc <- data_frame(id = metadata$dataset$`_id`$`$oid`,
			desc = metadata$dataset$description)
			nasa_desc %>%
			select(desc) %>%
			sample_n(5)

Onnowpurbo: Created page with " library(jsonlite) metadata <- fromJSON("https://data.nasa.gov/data.json") names(metadata$dataset) class(metadata$dataset$title) class(metadata$dataset$description) cla..."

2018-11-08T01:01:19Z

Created page with " library(jsonlite) metadata <- fromJSON("https://data.nasa.gov/data.json") names(metadata$dataset) class(metadata$dataset$title) class(metadata$dataset$description) cla..."

New page

library(jsonlite)
metadata <- fromJSON("https://data.nasa.gov/data.json")
names(metadata$dataset)

class(metadata$dataset$title)
class(metadata$dataset$description)
class(metadata$dataset$keyword)

library(dplyr)
nasa_title <- data_frame(id = metadata$dataset$`_id`$`$oid`,
title = metadata$dataset$title)
nasa_title

==Pranala Menarik==

* [[R]]