DFM 객체 정돈하기

install.packages("tm")
install.packages("topicmodels")
install.packages("tidyverse")
install.packages("tidytext")
install.packages("quanteda")
install.packages("scales")
library(tm)
library(topicmodels)
library(tidyverse)
library(tidytext)
library(quanteda)
library(scales)

 

data("data_corpus_inaugural")
data_corpus_inaugural

inaug_dfm <- quanteda::dfm(data_corpus_inaugural, verbose = FALSE)
inaug_dfm

 

inaug_tidy <- tidytext::tidy(inaug_dfm)
inaug_tidy

 

inaug_tidy %>%
tidytext::bind_tf_idf(term, document, count) %>%
dplyr::arrange(desc(tf_idf)) -> inaug_tf_idf

 

inaug_tidy %>%
tidyr::extract(document, "year", "(\\d+)", convert = TRUE) %>%
tidyr::complete(year, term, fill = list(count = 0)) %>%
dplyr::group_by(year) %>%
dplyr::mutate(year_total = sum(count)) -> year_term_counts

 

year_term_counts %>%
dplyr::filter(term %in% c("god", "america", "foreign", "freedom")) %>%
ggplot2::ggplot(mapping = aes(x = year, y = count / year_total)) +
ggplot2::geom_point() +
ggplot2::geom_smooth() +
ggplot2::facet_wrap(~term, scales = "free_y") +
ggplot2::scale_y_continuous(labels = scales::percent_format()) +
ggplot2::ylab("% frequency of word in inaugural address")

 

[출처] R로 배우는 텍스트마이닝, 줄리아 실기/데이비드 로빈슨 지음, 박진수 옮김, 제이펍, p89~92