Record Ordering Heuristics for Disclosure Control through Microaggregation

Statistical disclosure control (SDC) methods reconcile the need to release information to researchers with the need to protect privacy of individual records. Microaggregation is a SDC method that protects data subjects by guarantying k-anonymity: Records are partitioned into groups of size at least k and actual data values are replaced by the group means so that each record in the group is indistinguishable from at least k-1 other records. The goal is to create groups of similar records such that information loss due to data modification is minimized, where information loss is measured by the sum of squared deviations between the actual data values and their group means. Since optimal multivariate microaggregation is NP-hard, heuristics have been developed for microaggregation. It has been shown that for a given ordering of records, the optimal partition consistent with that ordering can be efficiently computed and some of the best existing microaggregation methods are based on this approach. This paper improves on previous heuristics by adapting tour construction and tour improvement heuristics for the traveling salesman problem (TSP) for microaggregation. Specifically, the Greedy heuristic and the Quick Boruvka heuristic are investigated for tour construction and the 2-opt, 3-opt, and Lin-Kernighan heuristics are used for tour improvements. Computational experiments using benchmark datasets indicate that our method results in lower information loss than extant microaggregation heuristics.