Non - interactive data release for distributive queries with differential privacy

Differential privacy is emerging as a strong notion for protecting individual privacy in privacy preserving data analysis or publishing. While it has been proven that it is possible to non-interactively release a database that are useful for all queries satisfying certain constraints with differential privacy, there is still a lack of general and efficient methods for achieving the data release. We propose a novel and efficient method for non-interactive data release for distributive queries with differential privacy. We examine queries with unlimited and limited output domain respectively and propose different differential privacy mechanisms which simultaneously guarantees differential privacy and the usefulness of released data. We also instantiate the general method for different queries including interval queries, parity queries and max queries. In addition to providing formal proofs of the differential privacy and usefulness guarantees of the released data using our method, we also present a set of experimental results and demonstrate the actual feasibility and performance of our method for different queries.