Skip to content
Streaming analysis of tweets using RStorm
HTML Other
  1. HTML 99.0%
  2. Other 1.0%
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


This repository contains code used to prototype a streaming framework for analysing Tweets about Comcast using the RStorm packages.

This is primarily being created for a presentation/tutorial at Interface 2015.


The purpose of the stream is to simulate/prototype a streaming framework for a someone wishing to analyze tweets matching a given topic, in this case Comcast. The stream will:

  1. Keep track and visualize (with a wordcloud) common terms associated with the topic.
  2. Classify and visualize the polarity (positive/negative/neutral) of tweets and visual common words in each class.
  3. Keep track of the proportion of positive tweets as time goes on.
  4. Visualize and report the rate of tweets in a given time frame.


This project requires several R packages:

  • RStorm
  • twitteR
  • sentiment*
  • wordcloud
  • dplyr
  • tidyr

Most of these packages can be installed from CRAN, but sentiment and one of its dependencies need to be installed from source:

	             repo = NULL, type = "source")
	             repo = NULL, type = "source")


Current Contents:

File Description
comcast_dash Folder containing shiny demo
presentation Folder containing presentation source code
tutorial Folder containing tutorial source code
stormr folder containing Storm package example
twitteRStorm.R Standalone R Script of tutorial. Readme file for project. License document for project.

Viewing the Dashboard

The dashboard is designed to prototype how a dashboard for monitoring tweets might took. You can run the app locally by calling shiny::runApp() in R from within the comcast_dash directory.

Alternatively, you can run the app from using the link:

You can’t perform that action at this time.