Description
Integrating data from multiple datasets is critical for many biomedical applications such as patient matching and population health research. Yet, privacy concerns pose major challenges during the interactive record linkage process. We propose the k-Anonymized Privacy Risk (KAPR) score that can monitor and quantify the actual privacy risk that occurs during this process and demonstrate its utility in a privacy enhanced interactive record linkage system called MINDFIRL (MInimum Necessary Disclosure For Interactive Record Linkage).
Learning Objective: After participating in this session, the learner should be better able to:
1. Learn properties required to measure the actual privacy risks due to the information disclosed during the interactive record linkage process
2. Learn how the k-Anonymized Privacy Risk (KAPR) score is calculated via an example
Authors:
Qinbo Li, Texas A&M University
Adam D'Souza (Presenter)
University of Calgary
Mahin Ramezani, Texas A&M University
Cason Schmit, Texas A&M University
Hye-Chung Kum, Texas A&M University
Presentation Materials: