The Tensorbros present...VACI-DATE

I know there is so much confusion about who can qualify for a vaccine and when. But don't worry, VACI-DATE is here! VACI-DATE helps people categorize themselves into groups to determine qualification.

Comment 1

The true HackDavis 2021 prizes were the friends and code we made along with the way

Inspiration

When we heard the COVID-19 vaccine was being distributed, we were ecstatic to return to normal life. However, after reading loads of news articles and local country guidelines, it felt impossible to know when our vaccination date would come. We designed VACI-DATE for you to know when it's your turn to get vaccinated!

What it does

It takes in information about the user and informs them what group of vaccination they are in. If it’s their turn, we show a map of local vaccination areas near them. If not, we ask the user for their phone number and we will text them when their turn is coming. We also use machine learning to predict the number of vaccinations to be administered when the user will be vaccinated.

Why we built it

COVID-19 has greatly disrupted our daily lives. It’s great news that a vaccine has been made and is being actively administered, but there aren’t a lot of sources out there that let a person know when they’re going to be vaccinated or what group they’re in. We built this tool as an easy way to help Californian residents know what phase of the vaccination plan they’re in, when they’ll be vaccinated, how many vaccines will be administered by then, and when they can get back to their normal life.

How we built it

We used React for our frontend with the axios library to make calls to our Firebase functions backend. Using Firestore with Firebase functions, we save our users’ phone numbers so that later, when it’s their turn, we can text them using the Twilio API. For predicting the number of vaccinations, we used this Google Colab to read in data and train the model. Libraries included pandas for data manipulation and Facebook’s Prophet for the forecasting machine learning model.

Challenges we ran into

  • Finding a dataset to train the forecasting model on
  • Figuring out ⭐s t a t e m a n a g e m e n t⭐
  • Figuring out where to use our DoorDash codes

Accomplishments that we're proud of

  • All of us stuck to the idea and managed to make something work
  • Making a website that uses libraries we’re not familiar with

What we learned

  • How to apply time series forecasting
  • How machine learning models were made to forecast COVID-19 infection rates
  • TEAMWORK MAKES THE DREAM WORK
  • How many amazing restaurants DoorDash has

What's next for VACI-DATE

  • Working with other states in the US
  • Fine-tuning forecasting model
  • Being bought out by Google

Built With

Share this project:

Updates