Blocking anonymized data
Document typeConference report
Rights accessOpen Access
Nowadays, privacy is an important issue, for this reason many researchers are working in the development of new data protection methods. The aim of these methods is to minimize the disclosure risk (DR) preserving the data utility. Due to this, the development of better methods to evaluate the DR is an increasing demand. A standard measure to evaluate disclosure risk is record linkage (RL). Normally, when data sets are very large, RL has to split the data sets into blocks to reduce its computational cost. Standard blocking methods need a non protected attribute to build the blocks and, for this reason, they are not a good option when the protected data set is completely masked. In this paper, we propose a new blocking method which does not need a blocking key to build the blocks, and therefore, it is suitable to split fully protected data sets. The method is based on aggregation operators. In particular, in the OWA operator.
CitationNin, J.; Torra, V. Blocking anonymized data. A: International Summer School on Aggregation Operators. "4th International Summer School on Aggregation Operators". Ghent: 2007, p. 83-87.