The effect of the Bootstrap method on additive fixed data perturbation in statistical database

In the information age, more and more data are stored in databases. Some data are retrieved in a statistical format as a means of protection. A statistical database provides summarized statistics to users while usually sheltering individual information. However, it has been found that users can send legal queries and deduce unauthorized information by recomposing queried data. Fortunately, the data perturbation technique can be used to improve the security of a statistical database by inserting minor biases into the database. This study uses Bootstrap method to demonstrate the possibility of deducing detailed information from individual data in a statistical database based on small samples, and then goes on to search for an effective perturbation distribution to ensure data security.