In HumanitiesDataAnalysis/TEIdy: Work with TEI and other XML Data as Tidy Text.

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(tidyverse)
library(tidytext)
library(TEIdytext)
fname = system.file("extdata", "frus1946v06.xml", package = "TEIdytext")
#fname = system.file("extdata", "Tomorrow_and_tomorrow.xml", package="TEIdytext")
#fname = system.file("extdata", "Mac.xml", package="TEIdytext")

all = TEIdy(fname, ignore = c("c", "w"), discard = "teiHeader") 

counts = all %>% 
  filter(div.document.subtype=="historical-document") %>%
  group_by(div.document.n) %>% 
  lift_data(persName.from, plaintext, name_from) %>%
  select(name_from, div.document.n, plaintext) %>%
  unnest_tokens(word, plaintext) %>%
  group_by(name_from, word, div.document.n) %>%
  filter(name_from != "") %>%
  summarize(count = n()) %>% arrange(-count) %>% 
  ungroup %>%
  group_by(div.document.n)




j = data %>% group_by(name_from) %>% grouped_llr(word,count)
j %>% arrange(-dunning) %>% 

HumanitiesDataAnalysis/TEIdy documentation built on May 11, 2019, 7:25 p.m.