How much privacy? — a system to safe guard personal privacy while releasing databases

Abstract

We propose two models to quantitatively measure the degree of privacy invasion based on the granular computing methodology. The total cost model measures the privacy invasion in light of the effort needed for an investigator to find individual’s private information.The average benefit model measures the privacy invasion in light of the benefit an investigator gets when his investigation improves the assessment of individuals private information.These two models can remedy the inadequacy of the deterministic formulation of privacy proposed in [4].These two measurements have been implemented in CellSecu 2.0, and a more relaxed generalization procedure, called external generalization, has also been implemented.