Within the challenge, our project performed very well because, through consolidated data and using the resources provided, the team managed to develop an idea whose social impact is very strong, proposing a solution for the current scenario of humanity, the coronavirus.
Collections, cleaning and structuring of databases were carried out which will be relevant to determine the most impacting correlation factors that describe the curve of cases of people infected by COVID-19. With 'Cases by Counties' data, graphs were generated that describe the number of cases in time by county. In addition, it was possible to apply regression models for the Alachua counties, which obtained an accuracy of 98.5% for the linear regression model and 99.2% for the polynomial regression model, and Miami-Dade, which obtained an accuracy of 97.9% for the linear regression model and 98.2% for the polynomial regression model. This, with the intention of developing predictive models for the cases of COVID-19, predictions which could be justified by the correlation factors that were chosen, once the project continues to evolve. Still, the prediction model that would possibly help in decision making, can be further improved by using resources such as SEDAC, Earth Data and Black Marble, which would add new correlation factors, illuminating the possible causes of incidence and prevalence of the pandemic.
https://youtu.be/BEuhOo0erVw
EARTH DATA, BLACK MARBLE and SEDAC.