In this note, we describe a method of simulated annealing for producing k-anonymity. For analytic purposes, there
is no reason to expect that the method will be superior or worse than the method of applying genetic algorithms used
by Iyengar (2002). The main appeal of simulated annealing is the amount of control it allows of the microaggregation
process. The k-anonymity problem is known to be NP complete.
Source: U.S. Census Bureau, Statistical Research Division