"Comparing Data While Keeping It Private"

Tanmay Ghai, a research engineer in the Networking and Cybersecurity Division at the University of Southern California's Information Sciences Institute (ISI) and a recent ISI alumnus, won the Viterbi Master's Student Award for Best Research in the Computer Science Department for 2022. Ghai conducted research on how to keep data private while resolving various entities and identifying relationships between them across datasets. For example, if a hospital and bank both have databases containing an individual's health and financial records, the task of understanding and linking those records that refer to that person is known as "entity resolution." It is a challenge, made even more difficult by the fact that this type of data is frequently highly sensitive. Ghai explained that comparing data between a hospital and a bank would necessitate them sharing the data so that the comparison could be performed, thus posing a privacy concern because the information is leaking from one entity to another. While it may just be a "name" or "username" in some cases, in more complicated scenarios, it could be an address, Social Security number, or even a bank account number, among other things. Maintaining the privacy of highly sensitive data increasingly complicates entity resolution because the data must be obfuscated to preserve privacy, making similarity comparisons difficult and expensive. This is particularly true for "fuzzy" or "approximate" matching, which takes into account differences in naming conventions and formats. Ghai and his co-authors presented "AMPPERE: A Universal Abstract Machine for Privacy-Preserving Entity Resolution Evaluation." It is a computational model that employs similarity metrics and privacy tools. They demonstrated that two parties can perform entity resolution over their data without leaking sensitive information by implementing AMPPERE using two different privacy tools over real-world datasets. This article continues to discuss Ghai's work in maintaining privacy in entity resolution.

USC Viterbi reports "Comparing Data While Keeping It Private"