Key points are not available for this paper at this time.
Users may be able to compromise databases by asking a series of questions and then inferring new information from the answers. The complexity of protecting a database against this technique is discussed here.
Dobkin et al. (Thu,) studied this question.