bigram_network.R

library(ggraph)

# as before, separate the mission column by words, but this time not per word, but per 2 words.
ms_bi <- ms %>% unnest_tokens(bigram, mission, token='ngrams', n=2)
ms_bi

# split the bigram column into two columns
bigrams_separated <- ms_bi %>%
  separate(bigram, c("word1", "word2"), sep = " ")

# filter out stop words
bigrams_filtered <- bigrams_separated %>%
  filter(!word1 %in% stop_words$word) %>%
  filter(!word2 %in% stop_words$word)

# count the bigrams
bigram_counts <- bigrams_filtered %>% 
  count(word1, word2, sort = TRUE)

bigram_counts

# filter to only keep bigrams above 1 
bigram_counts <- bigram_counts %>% filter(n > 1)

#create the bigram network graph

bigram_graph <- bigram_counts %>%
  igraph::graph_from_data_frame()

a <- grid::arrow(type = "closed", length = unit(.15, "inches"))

ggraph(bigram_graph, layout = "fr") +
  geom_edge_link(aes(edge_alpha = n), show.legend = FALSE,
                 arrow = a, end_cap = circle(.07, 'inches')) +
  geom_node_point(color = "lightblue", size = 5) +
  geom_node_text(aes(label = name), vjust = 1, hjust = 1) +
  theme_void()

#save the network
ggsave('mission_text_bigram_more_than_1.png', path=mypath)