How statistical noise is protecting your data privacy | Microsoft | The MarTech Alert | Scoop.it
Differential privacy is a technology that allows the collection and sharing of data while safeguarding individual identities from being revealed. Other privacy techniques can be limiting and can result in sensitive information, such as bank details, becoming discoverable.

Differential privacy introduces statistical noise – slight alterations – to mask datasets. The noise hides identifiable characteristics of individuals, ensuring that the privacy of personal information is protected, but it’s small enough to not materially impact the accuracy of the answers extracted by analysts and researchers. This precisely calculated noise can be added at the point of data collection or when the data is analyzed.

Before queries are permitted, a privacy “budget” is created, which sets limits on the amount of information that can be taken from the data. Each time a query is asked of the data, the amount of information revealed is deducted from the overall budget available. Once that budget has been used up and further information would then risk personal privacy being compromised, additional queries are prevented. It’s effectively an automatic shut-off that prevents the system from revealing too much information.