Search code examples
rtidytext

Mining financial articles R


I'm working on mining some financial articles using tidytext, I download the data from Reuters but then when I'm trying to turn each corpus into a data frame I get some errors about unnest command not taking functions as input...

Do you have any alternatives to get this into a tibble?

library(tm.plugin.webmining)
library(purrr)
company <- c("Microsoft", "Apple", "Google", "Amazon", "Facebook",
             "Twitter", "IBM", "Yahoo", "Netflix")

symbol <- c("MSFT", "AAPL", "GOOG", "AMZN", "FB", "TWTR", "IBM", "YHOO", "NFLX")

download_articles <- function(symbol) {
  WebCorpus(ReutersNewsSource(paste0("NASDAQ:", symbol)))
}

stock_articles <- data_frame(company = company, symbol = symbol) %>%
  mutate(corpus = map(symbol, download_articles))

stock_articles

stock_tokens <- stock_articles %>%
  unnest(map(corpus, tidy)) %>%
  unnest_tokens(word, text) %>%
  select(company, datetimestamp, word, id, heading)
stock_tokens

Solution

  • What's happening here is that some of the services have been deprecated, unfortunately, and tm.plugin.webmining is out of date. You can read some more details here. We are looking for a replacement dataset for this part of our book, but in the meantime, if you would like to explore using this code, I would recommend stripping down to just, say, 4 companies that appear to still be working.

    symbol <- c("MSFT", "AAPL", "AMZN", "IBM")