Amazon Web Services’ (AWS) machine learning team has released an open-source machine learning toolkit that helps predict the spread of COVID-19. The toolkit consists of multiple models specific to different regions as well as datasets. This will allow researchers and data scientists to better model the spread of COVID-19 in a community.
The toolkit also allows you to visualize how different interventions will potentially mitigate the spread of the virus based on previous data. You can see the structure of their model below.
The toolkit also gives examples of countries like France, Italy, and the US and how the spread of the virus could have been impacted by having stricter intervention measures in place. It also allows us to predict future spread. Say we are expecting a second or third wave of COVID-19 soon. You can use the model to see how the virus will potentially play out with different sets of preventive measures allowing you to get the best prevention policy in place.
The toolkit is more advanced than current models as it takes into account the spread from asymptomatic individuals as well.
The AWS team explains, “Our solution first tries to understand the approximate time to peak and expected case rates of the daily COVID-19 cases for the target entity (state/country) by analysis of the disease incidence patterns. Next, it selects the best (optimal) parameters using optimization techniques on a simulation model. Finally, it generates the projections of daily and cumulative confirmed cases, starting from the beginning of the outbreak [to] a specified length of time in the future.”
If you are looking to build a model based on the toolkit, you can access it here.
Image Source: NYSenate
