0

Is there a way to implement differential privacy on a Pandas data frame comprising non-numeric entries? Specifically, if I input the Pandas data frame, can I get a new data frame which is differentially private?

I am aware of a couple of packages for differential privacy, but not sure how to go about it. Any suggestions/directions shall be much appreciated.

Thanks!

Nanda
  • 361
  • 1
  • 5
  • 14
  • 1
    Have a look on : https://stackoverflow.com/questions/48008334/anonymize-specific-columns-with-pii-in-pandas-dataframe-python – Anurag Dabas Apr 22 '21 at 05:06
  • Thanks for the pointer. However, the methods in that link are very ad hoc. I was looking for a way to implement differential privacy on the pandas data frame. – Nanda Apr 22 '21 at 05:15

0 Answers0