Score:3

Anonymization Standards

tl flag

I'm interested in data anonymization techniques and scores for the quality of data anonymization in "the real world". I was told this is also part of cryptography.

What I know: In theory one can use k-anonymity, l-diversity or t-closeness to define the quality. Also the open gda score could be used. Common techniques are generalization, synthetic data and deleting data. I also studied the laws and found out, that there are no strict rules which state a standard for anonymization (I only found out, that the EuGH once said that reidentification should be "hard")

My question: How is anonymization done in practice? What standards are common? Which scores for quality are used and why?

Ievgeni avatar
cn flag
Anonymity doesn't make sense without any context. I think you should precise the use-case.
Titanlord avatar
tl flag
The main context I found in the literature is data in tabular form. For example csv or xlxx files. I don't mean anonymization done by VPN, Proxies, the Onion network or anything like that. I think the correct term for what I'm interested in is "data anonymization" (I added that to the question)
Paul Uszak avatar
cn flag
And/or [differential privacy?](https://crypto.stackexchange.com/questions/tagged/differential-privacy)
Ievgeni avatar
cn flag
@PaulUszak Is it a standard?
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.