Challenge 9 - Simple Anonymization - Solution

0 ×

You would like to post a question on the KNIME forum, but you have confidential data that you cannot share. In this challenge you will create a workflow which removes (or transforms) any columns that reveal anything confidential in your data (such as location, name, gender, etc.). After that, you should shuffle the remaining columns' rows such that each numeric column maintains its original statistical distribution but does not have a relationship with any other column. Rename these columns as well, such that in the end of your workflow they do not have any specific meaning. Let's see an example:

Example:
BEFORE ANONYMIZATION
Row Name Fav_Num Muscle_Mass
0 Victor 7 10
1 Aline 3 20
2 Scott 42 30

AFTER ANONYMIZATION
Row column column (#1)
0 3 30
1 42 10
2 7 20

Feel free to see our resources on data anonymization for inspiration (https://www.knime.com/blog/data-anonymization-in-knime-a-redfield-privacy-extension-walkthrough) , but note that the task here is much simpler! For reference, our solution only uses 7 nodes to anonymize and 3 additional nodes to do make sure the data truly was anonymized.

URL: Redfield http://redfield.se/

Nodes

• Row Filter2 ×
• CSV Reader1 ×
• Column Filter1 ×
• Column List Loop Start1 ×
• Column Rename (Regex)1 ×
• Loop End (Column Append)1 ×
• Table Difference Finder1 ×
• Table Row to Variable1 ×
• Target Shuffling1 ×

Extensions

• FeatureKNIME Base nodes

Links

• No links available