Disease Disaster Relief

• 

  The CDC publicly reports influenza outbreaks in the United States

• 

  Our initial analysis shows a tweet which seems to be impacted by recent flu-like or sickness activity

Inspiration

A while ago we explored sentiment analysis when it came to weather, but after a refactor, we can explore sentiment against being ill or sick. In addition, when we take into account the weather and the CDC's reported influenza outbreak, we can see if any interesting or emergent patterns exist.

What it does

Our website takes in data from Mapbox, Accuweather, Twitter, and the CDC to create a live visual map of possible outbreak candidates.

How we built it

We built it using our research but applied with HackPSU sponsors in mind.

Challenges we ran into

The AccuWeather API is rather cumbersome and can be hard to get the data that was needed to perform the analysis.

Accomplishments that we're proud of

We're most certainly proud of the integration with the CDC's datasets

What we learned

JS, Machine Learning, Live Scripting

What's next for Disease Disaster Relief

Not really sure, perhaps the code can again be refactored and retrained for some other interesting use. Data analysis can always lead to some interesting observations when put to a new angle.

Built With

• dataset
• html
• javascript
• learning
• machine-learning
• mapbox
• unsupervised