A general additive data perturbation method for database security

The security of organizational databases has received considerable attention in the literature in recent years. This can be attributed to a simultaneous increase in the amount of data being stored in databases, the analysis of such data, and the desire to protect confidential data. Data perturbation methods are often used to protect confidential, numerical data from unauthorized queries while providing maximum access and accurate information to legitimate queries. To provide accurate information, it is desirable that perturbation does not result in a change in relationships between attributes. In the presence of nonconfidential attributes, existing methods will result in such a change. This study describes a new method (General Additive Data Perturbation) that does not change relationships between attributes. All existing methods of additive data perturbation are shown to be special cases of this method. When the database has a multivariate normal distribution, the new method provides maximum security and minimum bias. For nonnormal databases, the new method provides better security and bias performance than the multiplicative data perturbation method.