Ninghui Li, Privacy Notions for Data Publishing and Analysis
Data collected by organizations and agencies are a key resource intoday's information age. The use of sophisticated data mining techniquesmakes it possible to extract relevant knowledge that can then be used for avariety of purposes, such as research, developing innovative technologiesand services, intelligence and counterterrorism operations, and providinginputs to public policy making. However the disclosure of those data posesserious threats to individual privacy. In this talk, we will present the evolvement of privacy notions fordata publishing and analysis, leading to our proposed membership privacyframework, which formalizes the intuition that privacy means that theadversary cannot significantly increasing its ability to conclude that anentity is in the input dataset. We show that several recently proposedprivacy notions, including differential privacy, are instantiations of themembership privacy framework, and that the framework provides a principledapproach to developing new privacy notions under which better utility can beachieved than what is possible under differential privacy. About the speaker: Ninghui Li is a Professor of Computer Science at Purdue University. Hisresearch interests are insecurity and privacy. Prof. Li is currently Vice Chair of ACM SpecialInterest Group on Security, Audit and Control (SIGSAC) and Program Chair of 2015 ACM Conference on Computer andCommunications Security (CCS). He is on the editorial boards of IEEE Transactions on Dependable and SecureComputing, Journal of Computer Security, and ACM Transactions on InternetTechnology.
Create your
podcast in
minutes
It is Free