A Toolkit for Multi-Faceted Geospatial Data Protection in Transactional and Analytical Processing using Searchable Encryption and Differential Privacy

Event Start
Event End
Building 9, Room 2322, Hall 1


The mobile revolution of the past decade led to the ubiquitous presence of location data in all application domains, ranging from public safety and healthcare to urban planning, transportation and commercial applications. Numerous services rely on location data to provide customized service to their users. At the same time, there are serious concerns with respect to protecting individual privacy, as location traces can disclose sensitive details to an untrusted service.
This talk will explore several research results in the area of location data protection under a broad spectrum of use case scenarios (e.g.., transactional and analytical queries, machine learning) and protection models (searchable encryption, differential privacy). Specifically, it will address: (1) supporting operational tasks such as secure alert zones using searchable encryption; (2) answering skyline queries on encrypted data; (3) combining DP and machine learning (ML) to support spatial queries and location-based recommendations; and (4) searchable encryption approaches to protect geospatial data on the blockchain.


Dr. Gabriel Ghinita is an Associate Professor at University of Massachusetts, Boston. Recently, he spent his sabbatical year (2018/19) as a Visiting Associate Professor at University of Southern California. Prior to joining UMB in Fall 2011, he was a Research Associate affiliated with the Purdue Cyber Center and the Purdue Center for Education and Research in Information Assurance and Security (CERIAS). He also held several visiting scholar appointments with Hong Kong University, City University of Hong Kong, and Nanyang Technological University, Singapore. Dr. Ghinita's research focuses on data security and privacy, with emphasis on protecting geospatial data. His earlier work published in ACM SIGMOD 2008 was the first to support practical nearest-neighbor queries with cryptographic-strength protection, and has close to 1000 citations to date (according to Google Scholar). His work on protecting location privacy received an Outstanding Paper Award at the ACM SIGSPATIAL 2009 conference, and a Distinguished Paper Award at the 2014 ACM Conference on Data and Application Security and Privacy (CODASPY). Dr. Ghinita served as Associate Editor for IEEE Transactions on Dependable and Secure Computing (TDSC) and as PC chair for the ACM Conference on Data and Application Security and Privacy (CODASPY). He serves regularly as reviewer for top journals and conferences such as IEEE TPDS, IEEE TKDE, IEEE TMC, IEEE TDSC, IEEE TIFS, ACM TODS, VLDBJ, PVLDB and IEEE ICDE.


Contact Person