For some supervised learning study I'd like to find a data set that show the impact of the current coronavirus crisis on some behavior. To be clear I am not looking for data about the health crisis (contaminated / ill / dead / vaccinated), but about behavior that were modified by the crisis. The goal is to make a small study of different metrics and how they behaved trough time.
It may be a bit vague, but I am looking for a data set with the folowing caracteristics :
- Tabular data
- Ideally clean / relatively Big
- Supervised framework (an ensemble of feature X, sampled trough time, and a target y, ideally some behavior observed over a given time horizon)
- Ideally a data set where the crisis can be seen similarly in both X and y, that is a 'deteriorating' X lead to a 'deteriorating' y. I'd like to avoid a data set where counter measures gave counterintuitive variations for y.
Anything in mind ? Or do I have to be more specific ?