PollTrack is a Twitterbot, written in Python, designed to track the polls during the 2016 presidential election.
Utilizing public state and national polling data from the Huffington Post Pollster API and the Twitter API, PollTrack collected and tweeted the latest polls using a dedicated Twitter account: @PollTrackBot.
Additionally, PollTrack could be configured to calculate a 7-day rolling average of national and state polls, and tweet the results along with a plot at a desginated time daily.
Note: PollTrack was designed to run using poll data released during the 2016 election season. The code was not designed to be run after the November 8 election.
To install simply clone the repository into a desired directory:
git clone https://github.com/haaspt/PollTrack.git
PollTrack requires the user to provide valid credentials for Twitter account, which is used to tweet new polls and graphs. To obtain credentials register a new app with Twitter's Application Manager. Once registered, edit the credentials.config.sample file in the root directory using your prefered text editor and enter your access tokens and consumer keys. Once edited, save the file as credentials.config.
The user can configure which averages are calculated and when they are tweeted by editing the plots.config file.
PollTrack runs in a continuous loop, it is ideally hosted on a server, or a machine with uninterrupted internet access. To run PollTrack simply type:
python main.py
PollTrack can also be run in the background using the following command:
sudo python main.py &