HiTaS: A Heuristic Approach to Cell Suppression in Hierarchical Tables

Cell suppression problem: A genetic-based approach

“…The HiTaS method [18] was developed for tables with hierarchical structure. It is a top-down methodology to decompose a hierarchical table into subtables, and then solving the CSP for each subtable.…”

Statistical Disclosure Control in Tabular Data

“…Although it is efficient, in practice tends to oversuppress cells and, moreover, it does not guarantee a feasible solution (indeed, it finishes with some underprotected cells). Some of the above drawbacks are also shared by the other heuristic, named Hitas [19]. That approach decomposes any table in a tree of smaller two-dimensional subtables and locally protects them by the previously cited optimal Benders decomposition approach.…”

“…However, they can be run with the same table from the τ-Argus package, which implements four of them: the optimal approach of [24], the shortest paths heuristic of [9], and the two (infeasible) heuristics of [19] and [25]. To compare them, in [26] a toy table 1H2D was generated with τ-Argus.…”

How to Make the τ-ARGUS Modular Method Applicable to Linked Tables

“…The modular approach for hierarchical table cell suppression (also called HiTaS, see [2] for a detailed description) subdivides a hierarchical table T into the corresponding set S(T) of simple, 'unstructured' linked (sub-)tables. The cell suppression problem is solved for each subtable separately.…”

“…(2) Compute a partial suppression pattern for subtable T a where only cells suppressed in (1) are eligible for (partial) suppression. (3) Assign the distances between the bounds of intervals given by the partial suppression pattern and the cell value of any suppressed cell s of T a as protection level to s when protecting any other subtable T b containing cell s.…”