Compromising anonymity in identity-reserved k-anonymous datasets through aggregate knowledge
Kevin De Boeck, Jenno Verdonck, Michiel Willocx, Jorn Lapon, Vincent Naessens · 2024
Data processors increasingly rely on external data sources to improve strategic or operational decision taking. Data owners can facilitate this by releasing datasets directly to data processors or doing so indirectly via data spaces. As data processors often have different needs and due to the sensitivity of the data, multiple anonymized versions of an original dataset are often released. However, doing so can introduce severe privacy risks.