k-Anonymity

A privacy property where every record in a dataset is indistinguishable from at least k-1 other records on quasi-identifier values. Prevents singling out individuals through quasi-identifier combinations. Introduced by Latanya Sweeney in 2002.