#6: Privacy-Preserving Record Linkage

On the 02-December-2021 at 11:00 a.m. the sixth lecture of the Living Lab lecture series took place. In this talk, Florens Rohde introduced the audience to state-of-the-art methods, challenges and his research topics in Privacy-preserving Record Linkage.

A crucial task in data integration is record linkage or entity resolution that aims at linking records that refer to the same  real-world entity, such as persons. For example, in multi-site medical  research data from several sources (e. g., hospitals) has to be matched to analyze correlations between medical data of the same patient in multiple data sources.
Typically, there is a lack of global identifiers, therefore the linkage can only be achieved by comparing available quasi-identifiers, such as name, address or date of birth. However, in many cases, data owners are only willing or allowed to provide their data for such data integration if there is sufficient protection of sensitive information to ensure the privacy of persons. Privacy-preserving Record Linkage addresses this problem by providing techniques to match records without revealing plaintext data to each other or to third parties.

You can rewatch this lecture on YouTube.