In quanteda/quanteda: Quantitative Analysis of Textual Data

knitr::opts_chunk$set(echo = TRUE)

require(quanteda)
quanteda_options(threads = 8)
sessionInfo()

corp <- readRDS("/home/kohei/Documents/Brexit/Data/data_corpus_guardian.RDS")

system.time(
    corp2 <- corpus_sample(corp, 5000)
)

system.time(
    toks <- tokens(corp)
)

system.time(
    toks2 <- tokens_sample(toks, 5000)
)

system.time(
    toks3 <- tokens_remove(toks, stopwords("en"))
)

system.time(
    mt <- dfm(toks)
)

system.time(
    mt2 <- dfm_group(mt, "date")
)

system.time(
    mt3 <- dfm_remove(mt, stopwords("en"))
)