Defferential Privacy

Definition:

Differential privacy is a formal mathematical model of privacy. In simplified terms, differential privacy requires that the output of an analysis should reveal (almost) no information specific to any individual within that dataset, where by “specific” we mean information that could not be inferred about the individual from statistical facts about other individuals

Commentary: