Scaling up a hybrid genetic linear programming algorithm for statistical disclosure control

Abstract

This paper looks at the real world problem of statistical disclosure control. National Statistics Agencies are required to publish detailed statistics and simultaneously guarantee the confidentiality of the contributors. When published statistical tables contain magnitude data such as turnover or health statistics the preferred method is to suppress the values of cells which may reveal confidential information. However suppressing these 'primary' cells alone will not guarantee protection due the presence of margin (row/column) totals and therefore other 'secondary' cells must also be suppressed. A previously developed algorithm that hybridizes linear programming with a genetic algorithm has been shown to protect tables with up to 40,000 cells, however Statistical Agencies are often required to protect tables with over 100,000 cells.This algorithm's performance highly depended on the choice of mutation operator so firstly this dependency was removed. As the algorithm is unable to protect larger tables due to the time it takes for its fitness function (a linear program) to execute a series of modifications have been applied. These modifications significantly reduced its execution time which in turn greatly extend the capabilities of the hybrid algorithm to the point that it can now protect tables with up to one million cells.