Quasi-identifiers (Quasi-dientifier, QI)

Quasi-identifierfrom Wikipedia, the free encyclopedia

quasi-identifiers is pieces of information that is not of the themselves unique identifiers, but is sufficiently w Ell correlated with the entity that they can is combined with other quasi-identifiers to create a unique identifier. ^[1]

Quasi-identifiers can thus, when combined, become personally identifying information. This process is called re-identification. As an example, LaTanya Sweeney have shown that even though neither gender, birth dates nor postal codes uniquely identify a n individual, the combination of all three are sufficient to identify 87% of individuals in the States. ^[2]

The term is introduced by Tore Dalenius in 1986. ^{[3] Since Then, Quasi-identifiers has been the basis of several attacks on release D data. For instance, Sweeney linked health records to publicly available information to locate the Then-governor of Massachusetts ' Hospital records using uniquely-identifying quasi-identifiers, [4] [5" and Sweeney, Abu and Winn used public voter records to re-identify participants in the P Ersonal genome Project. [6] Additionally, Arvind Narayanan and Vitaly Shmatikov made use of Quasi-identifiers to De-anonymize data released by Netflix. [7]}

Motwani and Ying warn about potential privacy breaches being enabled by publication of large volumes of government and bus Iness data containing quasi-identifiers. ^[8]

Quasi-identifiers (Quasi-dientifier, QI)