Review description of input dataset

[evolutionv2]

Please all check whether you are comfortable with the described input format, whether variables should be added or discarded or put as mandatory or not.

[evolutionv2]

You can either add comments here or make changes to the files itself if you are convinced.

[lukric]

General remarks

• I think we should ask/demand for uniform encoding of input.csv(special charactes, umlauts, accents) - UTF-8
• county_id, community_id, state_id: are all countries using NUTS levels? We should allow arbitrary IDs.
• age_group: do we allow arbitrary levels, or values from a given list?

Data necessity

Would we use the following variables for signal detection? Would we use them for (graphical) presentation in the shiny app? Of no, we should discuss removing them from the input.csv in order to minimize data.

• date_vaccination
• occupation
• place_of_infection
• hospitalization (should not be mandatory)
• death (should not be mandatory)
• vaccination (should not be mandatory)
• symptoms
• risks

If Austria was a piloting country we would not be able to provide this data for most diseases.

[mlbach]

That is some great feedback on the data necessity! That is exactly why we need to have the discussions, because we don't know what type of data can be provided by each country.

I also agree on the UTF-8 encoding. Regarding the age group, I think it mostly depends on whether or not data should be comparable between countries. If yes, then we should specify a list of values.