So I have been tasked to anonymize clinical data that..most of the time would be present in the form of tables.

I've looked into algorithms like generalization,redaction,K-anonymity, l-diversity,t-closeness. But I am not sure how to robustly implement them. I've also looked into Microsoft presidio.

Are there any better approaches for de-identification of tabular data? What's used in the industry now to anonymize a database or a table?

Thanks, any help would be appreciated.

  • 1
    $\begingroup$ I think this question is too broad to answer in this forum. There is an entire field of work on anonymization and one could write multiple books on the subject. This site is best for narrowly focused questions that can be answered concisely in a few paragraphs. Any votes from the community? $\endgroup$
    – D.W.
    Commented Mar 18 at 20:32


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.