We were stuck, lost, and exploring the vast and exotic world of APIs, when we came across one that provided us with a tally of the frequency with which certain prominent, transmissible diseases had been documented in the U.S. Endowed with such knowledge, we knew we had no choice, but to implement something that could really put its value into perspective.

What it does

Webidemic is a website that displays the density of a number of prominent diseases in the U.S. throughout the country's history. It elaborates on how the information gathered from this API could be relevant to current and future disease-spreading illustrates the changes in disease frequency with the passage of time and provides an insight into how we may predict the migrations of diseases in the future.

Challenges we ran into

We originally wanted to predict the migration of the most popular epidemics using Recurrent Neural Networks, a deep learning algorithm that has the ability to predict time series data. However, the data provided from the google cloud services and various government release datasets do not contain extra information other than the report date and the type of disease recorded. Due to the limitation of time, we decided to abandon the possibility of collecting more features, such as temperature and top news, about each report date.

What's next for Webidemic

Saving the world, one disease at a time.

Share this project: