Statistical offices are concerned with problems of protecting confidential information when publishing data in statistical tables. One method to avoid disclosure is the method of cell suppression, in which the values of the sensitive cells in the table are suppressed from publication. To prevent the values of the sensitive cells from being calculated from the tables totals, additional suppressions are necessary. The problem of minimizing the loss of information caused by the additional suppressions is a difficult optimization problem. We present and compare the performance of some heuristics for cell suppression for general three-dimensional tables.
|Publication status||Published - 1999|