Data anonymization is a process that takes data pertaining to certain users and removes identifying
information. One technique for data anonymization is called “k-anonymity.” In this technique, unique
identifiers are completely removed, and some data that could lead to identification (called quasiidentifiers)
are generalized (replaced with a less-specific value).
A data set is said to be k-anonymous if every query involving quasi-identifiers returns either zero or at
least k rows.