3 ms·
If you’re curious about the matter, there’s an alternative technique called differential privacy that adds randomization to the data, and is able to provide gua
by lambdasquirrel 8y ago
If you’re curious about the matter, there’s an alternative technique called differential privacy that adds randomization to the data, and is able to provide guarantees that individual records cannot be identified. k-anonymity is subject to e.g. linkage attacks of the kind that differential privacy seeks to eliminate.
- shawn 8y agoBonus: The differential privacy algorithm is so simple. "When you're about to insert some data, flip a coin. If it's heads, insert it as-is. If it's tails, insert a random value."
- barbegal 8y agoBut by adding random data, differential privacy destroys the ability to apply the usual statistical analysis techniques to the dataset.