Add up-to-date Corona Case Data to your Analyses [Version 2]

If you would like to add the number of Corona Cases worldwide to your analysis, you can use the Machine Learning App “COVID-19 Data Integration” which is readily available in your IBC team. The following how-to guide describes the necessary steps in detail.

Features / Data Scope:

  • per day and country: data about confirmed cases, deaths, recovered and active cases
  • delta values (difference to the day before)
  • population information per country
  • doubling time per day and country
  • updates data automatically in a predefined schedule
  • can be added to your data model on every two-digit or three-digit language code
  • standard analyes will be installed by the App and

Source: Johns Hopkins University (Terms of Use)

Install Machine Learning App
Go to the “Machine Learning” Tab and create a “New App”.
If you cannot access Machine Learning, reach out to your Customer Success Manager!

Select “COVID-19 Data Integration” as Application Type and click “Create”.
image

Change Permissions
Make sure to change the permissions for the Application Key according to the information given on the screen.
image

Select Data Pool and Data Model
Choose the Data Pool and the Data Model where the Covid-19 data should be used.
Afterwards, you can see a preview for the Covid-19 data.
image

Select Country column to join corona data on (optional)
Choose a column from any table in your Data Pool.
This column has to contain values in the form of ISO-2 Country Codes (for example “DE” or “US”).
You can also select several columns.
If you do not want to join the Covid-19 Data to any existing table, you can just skip that step.

Push Data & Create Scheduler
In the last step, you can change the name of the table, if necessary.
Click “Push results to Data Model and create Scheduler” to finalize the Covid-19 Data Integration.

------ Version 0.1 of the COVID-19 Script [DEPRECATED] ------

The HowTo-Guide on the following Dropbox-Link explains how to set-up the Covid-19 Live Data manually. This guide is deprecated - please use the Machine Learning App solution described above:

2 Likes

Hi, i’ve downloaded the files but it seems to be a little bit different between the country codes to make the joins.

country_codes.csv - CODE column is a 2 letter code
cases_time.csv - Doesn’t has a CODE column. It only has Country_Region column and cannot match with country codes. Not even by name because names are different
country_population.csv - Has a 3 letter country code that doesn’t match to any country_code on the country_code.csv file.

I don’t know if i am seeing a wrong file or if you can help making the adjustments.

Thanks!

Hello Erick,
thank you for your message! You are right - these column do not exist in the source files. However, by using the Transformation.sql (as a Transformation in your Data Pool), the tables are properly joined by using the Country Name. Unfortunately, this leads to some minor mismatches for some countries, where the Name is not 100% the same. But for most of them, the join works.
Detailed information on how to work with the files can be found in the HowTo above.

I hope this helps!

Hi, I don’t have the cloud version but the enterprise server edition 4.6. How can I manage with the set up described here in my case?

Thanks

Hi sgizm,

we just released a new version of the Covid-19 Data Integration Script. The previous guide is updated accordingly.

Unfortunately, the described solution is only possible for teams in the Intelligent Business Cloud. However, if you have a Python environment installed on your server, we can look for alternative ways of integrating it. In that case, please reach out to me personally (v.grundler@celonis.com).