697DB-final

There are two main code files in this repository:

get_and_prepare_tweets.Rmd

This is the code that uses the Twitter and Google Geocoding APIs to fetch and prepare data for the Shiny app. It produces a few stages of intermediary files along the way to the final data file called located_tweets_senti.rda, which contains the variable located_tweets_senti, a data frame with all of the Twitter columns, plus sentiment analysis columns and latitude and longitude returned by Google for textual location data scraped from the Twitter users' bios.

It requires a config.yml file with your API credentials, in the following format:

default:
  # Used to create the twitter token
  twitter_app : "yourvalue"
  twitter_consumer_key : "yourvalue"
  twitter_consumer_secret : "yourvalue"
  twitter_access_token : "yourvalue"
  twitter_access_secret : "yourvalue"
  
  # google API
  google_geocoding_apikey : "yourvalue"

The directory "data" will be created if it doesn't exist, and files are written to it. data/located_tweets_senti.rda is loaded by app.R.

app.R

This is the Shiny application. It loads the data/located_tweets_senti.rda file created by get_and_prepare_tweets.Rmd above.

You can see it running at https://stevelinberg.shinyapps.io/697DB_final_project/.

Steve Linberg
https://slinberg.net
steve@slinberg.net