COVID-19 has already left us all in the homebody-lifestyle for the sake of our safety, but sometimes a trip out is unavoidable. At those times you ask yourself; "How risky will it be going to ______?". We thought that creating an app to measure your potential risk of contracting COVID-19 in a given area would be helpful to those that want to be more wary of their surroundings.

What it does

Since this is an unfinished project, the functionality of the program is extremely limited. As of right now, it reads in the NYC COVID-19 dataset as well as the NYC train system dataset and plot the data points on a graph.

How we built it

We discussed the type of data we wanted and where to find it, then read the data into a python notebook to analyze and really try to make meaning with it.

Challenges we ran into

Finding the data was easy, but the real question was "What can we do with this?". We wanted to use machine learning to classify outcomes as "safe" or "unsafe", however we lacked the amount of quality data needed for that. It's available to find but we realized there is still a lot to learn before we can implement it the way we want to. It lead to a lot of research and many trials & errors. The process came with a few large bugs, and everyone knows how fun that is at 4AM. Things just did not work out the way we expected it to, and thats okay because we came here with the mindset to learn.

What we learned

We learned many things about the topic of machine learning, such as K-Means clustering for unsupervised learning. We also picked up many data analyses skills that could be used with programming. And most importantly, how to collaborate as a team.

What's next for COVID-19 Travel App

We want to continue this project outside of the scope of the hackathon, so we're going to make it an open sourced project. This way, we can find more data for other cities and hopefully, countries too. This would allow users from anywhere determine their risk of contracting COVID-19 while traveling or commuting to wherever they need to.

Built With

Share this project: