Inspiration

With the recent coronavirus craze, there is a lot of news from a lot of sources- which makes getting solid information about COVID that much harder. We knew there was a better way, so we put our minds together to make a dashboard that shows play-by-play updates of COVID 19 both in numbers and in the news.

What it does

This is an early prototype for sure - but with user input for their country/state, the dashboard shows a graph of past and current case numbers, and it would show a realtime number for the number of cases in that country or state. Keep in mind that this number is really real time - the number that you requested a minute ago might be hundreds of cases different than the number you get when you input that country again!

How I built it

We used Python's urllib and requests to crawl through various data sources(some which were, at the start, infuriatingly hard to get to), and then we used pandas to format it nicely, matplotlib to plot it, and flask to display it all.

Challenges I ran into

Some of the data sources were super dev-unfriendly, so it was hard to figure out how to get the data in the more efficient way possible without having to compromise user experience. It was also challenging to deal with the dataframes themselves, as they can be quite bulky and a bit hard to manipulate as times.

Accomplishments that I'm proud of

We overcame the data source access issue, and while we came into the competition a bit late, we are proud that we were able to recover and churn something out.

What's next for COVID-Dashboard

If we haven't by the time that you are reading this, we are going to add a news summary for your country or state, and we are also going to add dynamic counters for the realtime, play-by-play cases data that we are getting. We had also made a preliminary prediction backend for COVID based on previous data, but we ran into some roadblocks - that would also be something that we would be willing to pursue. Lastly, adding things like hyperlocal interactions - number of cases and recommended shops, hospitals, etc in your specific area - is something that we were really interested in.

Built With

Share this project:

Updates