Alexandros Bampoulidis, Ioannis Markopoulos, Mihai Lupu (2019)

Extensive research in de-anonymisation has shown that in datasets not containing any personally identifying information (PII)—name, address, etc.—individuals can be identified through quasi-identifiers (QIs)—attributes whose combination serves as a unique identifier. In order to deal with this issue, necessary anonymisation measures need to be taken which, however, reduce the quality of a dataset by modifying its values. Data publishers may deem that some QIs are more important than others and, therefore, should be distorted as little as possible in the anonymisation process. Most existing tools do not take such weighting into consideration. In this demo, we present a tool addressing this issue by utilising a local recoding algorithm for k-anonymity, capable of outperforming the existing state-of-the-art tool ARX.

Read the full paper here