R: trigram

library(dplyr)
library(tidytext)
library(janeaustenr)
library(tidyr)
library(igraph)
library(ggplot2)
library(ggraph)
library(readtext)

text <- readtext("out.txt")
text_trigrams <- text %>%
                unnest_tokens(trigram, text, token = "ngrams", n = 3)
text_trigrams
trigrams_separated <- text_trigrams %>%
separate(trigram, c("word1", "word2", "word3"), sep = " ")
# stopwords Indonesia
trigrams_filtered <- trigrams_separated %>%
       filter(!word1 %in% stopwords::stopwords("id", source = "stopwords-iso")) %>%
       filter(!word2 %in% stopwords::stopwords("id", source = "stopwords-iso")) %>%
       filter(!word3 %in% stopwords::stopwords("id", source = "stopwords-iso"))
trigram_counts <- trigrams_filtered %>%
      count(word1, word2, word3, sort = TRUE)
trigram_graph <- trigram_counts %>%
  filter(n > 20) %>%
  graph_from_data_frame()
trigram_graph
set.seed(2017)
ggraph(trigram_graph, layout = "fr") +
  geom_edge_link() +
  geom_node_point() +
  geom_node_text(aes(label = name), vjust = 1, hjust = 1)