Skip to content

wlz99/c3aiHackathon

Repository files navigation

CovidML – an operational tool to moderate the pace of the COVID-19 epidemic

visit https://covid-machine-learning.herokuapp.com/ submitted for C3.ai COVID-19 Grand Challenge workflow

Inspiration

The human cost of coronavirus has continued to mount, with more than 40.3m cases confirmed globally and more than 1.11m people known to have died. Building an intuitive dashboard visualization tracking the most updated status of Covid is beneficial to the public at large.

What it does

CovidML is a web application that visualizes the trend of COVID-19 worldwide with future spread prediction. It allows users to effectively visualize current and upcoming situations of the pandemic spread.

How We built it

Our web app is based on the Flask python framework. We obtained multiple datasets from c3.ai datalake. The visualizations on the plots page display the status quo of infection cases and fatalities across the world using Datawrapper tool. We produced geo-referenced plots using a combination of libraries. We used pandas to generate a data-frame for the dataset, and then tried different models to forecast the infection trends.

Challenges We ran into

Mining and analyzing the data to build the data visualizations was a hard step. Even with perfect reporting, there are fundamental delays in what such data can tell us. For example, new infections today reflect virus activity as of 9 to 13 days ago (which depends, in turn, on social distancing interventions from up to 17 days prior), as derived from the lag correlation analysis. Not factoring in such considerations have led to significant over-estimation of hospitalization needs nationwide.

What's next for CovidML

We hope to expand our web application to cover countries all around the world, add a page that visualized the multiplex impacts of Covid, and try more modeling techniques to forecast the future trend of the pandemic.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •