K-anonymity

K-anonymity is a theoretical measure of how anonymous a data set is, such that without additional data, what is the mathematical guarantee an individual in the data set cannot be directly-identifiable.