Reasoning under Uncertainty in On-Line Auditing.

We propose a Bayesian approach to reasoning under uncertainty in on-line auditing of Statistical Databases. A Bayesian network addresses disclosures based on probabilistic inferences that can be drawn from released data. In particular, we deal with on-line max and min auditing. Moreover, we show how our model is able to deal with the implicit delivery of information that derives from denying the answer to a query and to manage user prior-knowledge