Skip to content

About

Interactive Shiny App for downloading and processing data from Twitter's Search API using R + Shiny + R libraries

Resources

Stars

0 stars

Watchers

1 watching

Forks

Latest commit

 

History

4 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 

Repository files navigation

TwitterShinyApp

Interactive Shiny App developed by Carlo R. M. A. Santagiustina for downloading and processing data from Twitter's Search API using R and the following R libraries (shiny,Matrix,rjson,jsonlite,httr,rvest,tidyverse,shinyjs,DT,openxlsx,quanteda,rtweet,stringr,tibble,rtweet,shinythemes,ggplot2,quanteda). You can use the TwitterShinyApp to download up to 18000 tweets every 15min from the Twitter Search API (get-search-tweets endpoint), by keyword and/or by geocode (for more info on the endpoint see: https://developer.twitter.com/en/docs/tweets/search/api-reference/get-search-tweets.html). The TwitterShinyApp also allows you to analyse and process through a user friendly interface downloaded tweets. Implemented functionalities include:

  • Summary statistics by variable (column);
  • Ngrams and term/pattern context analysis;
  • Time series of tweets counts for multiple time frequencies (weeks, days, hours, mins, secs);
  • Textual data cleaning (text and user description);
  • Extraction of URLs;
  • Extraction of emoticones;
  • Extraction of hashtags (#), usertags (@) and fintags ($);
  • Hashtags and usertags frequency analysis and wordclouds for most frequent hashtags/usertags;
  • Hashtag and usertag cooccurrence network (construction + visualization) for most frequent hashtags/usertags;

To run the app locally:

  1. Download the app - nocredentials.R file;
  2. Install R version 3.5 (or later) and RStudio;
  3. Open the app - nocredentials.R file in RStudio;
  4. Install automatically the suggested packages in RStudio ( install.packages("shiny","Matrix","rjson","jsonlite","httr","rvest","tidyverse","shinyjs","DT","openxlsx","quanteda","rtweet","stringr","tibble","rtweet","shinythemes","ggplot2","quanteda") );
  5. Insert your Twitter Developer App credentials (see your Twitter developer account credentials at: https://developer.twitter.com/en/apps/ ) in the corresponding fields (see script lines 262, 264, 266, 268,270)
  6. Run the Shiny app using the Run App button (scripting window top-right corner);
  7. Enjoy!

Reminder: Twitter Search API limits allow up to 180 queries (18 000 tweets downloaded) every 15min.

Apache Licence 2.0

contact: carlo.santagiustina@unive.it

About

Interactive Shiny App for downloading and processing data from Twitter's Search API using R + Shiny + R libraries

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages