Larry on Fire

Installing the Application

1. Run bash install.sh in the top-level directory, this will install all required libraries in a virtual environment
2. Enter the virtual environment with source virtual_larry/bin/activate
3. Launch the dash server with ipython3 larry_on_fire.py and click on the link displayed in the output

Notes:

• Please run this application locally for optimal performance.
• When initially loading the application, it can take about a minute.
• When using the filters in the browser, the application takes a few seconds to reload, especially when loading tweets from the entire country.

Interacting with the Application

Once the three steps above are executed, one can interact with the application on the web browser by toggling three things:

• Year (dropdown menu)
• Season (button)
• Geography (button)

The prior two apply to all 4 visualizations, while the last filter only applies to the wordcloud and LDA. This allows the user to (1) locate the fires, (2) look at the tweet intensity compared to the wildfire intensity, and (3) get an overview of the discussed topics.

Interacting with the API

1. Run ipython3 api_interaction.py in the top-level directory
2. Enter a start and an end date in the command line (format: YYY-MM-DD)
3. View the output of the API in data/twitter_data/simulation.csv