A naukri.com initiative
Johndcook
1w
80
Deleting vs Replacing Names
- When deidentifying personal data, structured data can have real names deleted or replaced with random names without impacting privacy.
- For structured data in databases, removing a column with names is a necessary step for deidentification, but replacing names doesn't affect privacy much.
- In cases like software testing with patient data, replacing real names with randomly generated names can be useful.
- For unstructured data like doctor's notes, replacing names with synthetic names is better for privacy than just removing them.
- Replacing names in unstructured data helps maintain privacy as software might not accurately identify all names.
- Synthetic names obscure failures of data deidentification software better than 'NAME REDACTED' placeholder.
- Using synthetic names in unstructured text deidentification is preferable as software errors are less obvious.
- An effective process for synthesizing names should have a low error rate and be unbiased.
- Synthesizing names rather than removing them is better if the error rate is sufficiently low and unbiased.
- Deidentification software often has error rates, but it doesn't need to be zero to comply with privacy regulations like HIPAA.
Read Full Article
4 Likes
For uninterrupted reading, download the app