Additive Gaussian Noise Based Data Perturbation in Multi-Level Trust Privacy Preserving Data Mining

摘要

Data perturbation is one of the most popular models used in privacy preserving data mining. It is speciallyconvenient for applications where the data owners need to export/publish the privacy-sensitive data. Thiswork proposes that an Additive Perturbation based Privacy Preserving Data Mining (PPDM) to deal withthe problem of increasing accurate models about all data without knowing exact details of individualvalues. To Preserve Privacy, the approach establishes Random Perturbation to individual values beforedata are published. In Proposed system the PPDM approach introduces Multilevel Trust (MLT) on dataminers. Here different perturbed copies of the similar data are available to the data miner at different trustlevels and may mingle these copies to jointly gather extra information about original data and release thedata is called diversity attack. To prevent this attack MLT-PPDM approach is used along with the additionof random Gaussian noise and the noise is properly correlated to the original data, so the data minerscannot get diversity gain in their combined reconstruction.