Analyzing the Empirical Effectiveness of COVID-19 Vaccines

Inspiration

COVID-19 is one of the biggest challenges in the world right now and vaccines are one of the tools we can use to combat COVID-19. I wanted to see how the transmission rate of COVID-19 has been affected by vaccination and different COVID variants.

What it does

Creates visualizations, of the United States, of the distribution of COVID-19 vaccines, of each type of vaccine, of COVID statistics throughout different key times (before vaccines, different variants, etc.), and of the transmission rate history in Washington State.

Attempted to predict new COVID cases given population, census area, total cases, and with and without vaccination data.

How we built it

I wrote the Python code in the EdStem workspace.

Challenges we ran into

I was unable to accurately predict new COVID cases using either a DecisionTreeRegressor or a Neural Network. This might have been because I didn't take into account additional factors that can affect new cases such as state regulations and human behaviors.

Accomplishments that we're proud of

I am proud to have created cool visualization using COVID data from the CDC. I also believe that the results I found are relevant to the real world.

What we learned

I learned to import, clean, combine, and visualize real-world data using a variety of python libraries such as geopandas and matplotlib. I also experimented with decision trees and neural networks from the scikitlearn library.

What's next for Analyzing the Empirical Effectiveness of COVID-19 Vaccines

In the future, I would like to make models that better predict the number of new COVID cases with more input features. I also want to continue tracking COVID data in case things change (ex: a new variant).

Built With

• python