Created
September 15, 2022 09:09
-
-
Save cboulanger/a09b1908c3c5160b8e01870184e5fbe6 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| library(tidyverse) | |
| library(tidytext) | |
| library(xts) | |
| library(tstools) | |
| base_dir = "C:\\Users\\Boulanger\\ownCloud\\Abteilung 3\\Projekte\\Legal Theory Graph" | |
| corpus_dirs = c() | |
| for (dir in list.dirs(base_dir)) { | |
| if (dir != base_dir) { | |
| corpus_dirs = c(corpus_dirs, dir) | |
| } | |
| } | |
| corpora = list() | |
| for (dir in corpus_dirs) { | |
| setwd(dir) | |
| print(dir) | |
| name = basename(dir) | |
| if (! name %in% names(corpora)) { | |
| files = list.files(dir, pattern = ".txt$") | |
| # create a word list of all files in the directory | |
| df <- map_df(files, ~ data_frame(txt = read_file(.x)) %>% | |
| mutate(filename = basename(.x)) %>% | |
| unnest_tokens(word, txt)) | |
| # group by year and compute total number of words | |
| df <- df %>% | |
| mutate(year=str_extract(filename,"(19|20)\\d\\d")) | |
| print(paste("Years:", paste(unique(df$year), collapse = ", "))) | |
| df <- df %>% | |
| dplyr::count(year, word) %>% | |
| bind_tf_idf(word, year, n) %>% | |
| complete(year, fill = list(count = 0)) %>% | |
| group_by(year) %>% | |
| mutate(year_total = sum(n)) %>% | |
| mutate(year=as.numeric(year)) %>% | |
| mutate(date = as.Date(paste(year,"-01-01", sep = ""))) | |
| corpora[[name]] = df | |
| } | |
| } | |
| #search for key terms | |
| # absolute werte -> "n / year_total" weglassen | |
| xts2ts <- function(XD, freq) { | |
| maxRow <- nrow(XD) | |
| startYM <- c(.indexyear(XD[1]) + 1900, .indexmon(XD[1]) + 1L) | |
| endYM <- c(.indexyear(XD[maxRow]) + 1900, .indexmon(XD[maxRow]) + 1L) | |
| ts(as.numeric(XD), start = startYM, end = endYM, frequency = freq) | |
| } | |
| ts_list = list() | |
| for (name in names(corpora)) { | |
| df <- corpora[[name]] %>% filter(word %in% c("ehrlich")) | |
| ts <- seq.Date(as.Date("1970-01-01"), as.Date("2019-12-31"), by="year") | |
| df <- data.frame(date=ts) %>% full_join(df) | |
| xts_data <- xts(df$n, order.by=df$date, frequency = 1) | |
| ts_data = xts2ts(xts_data, 1) | |
| ts_list[[name]] <- ts_data | |
| } | |
| figure3 <- tsplot(ts_list, | |
| plot_title="'ehrlich' im Rechtssoziologiekorpus", | |
| left_as_bar = T, | |
| theme = init_tsplot_theme(bar_gap = 10)) | |
| figure3 |
Author
cboulanger
commented
Sep 15, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment