BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY: Privacy Challenges and Solutions for Data Sharing - Dr Aris Gkoul
 alas-Divanis\, IBM Research\, Dublin
DTSTART:20130213T141500Z
DTEND:20130213T151500Z
UID:TALK43290@talks.cam.ac.uk
CONTACT:Stephen Clark
DESCRIPTION:Data sharing to support research or other purposes can be high
 ly beneficial but may also entail serious\nprivacy risks when the shared d
 ata refers to individuals or can lead to the disclosure of sensitive knowl
 edge patterns\,\nsuch as trade secrets\, when mined. To overcome these thr
 eats\, the research areas of privacy-preserving data\npublishing and priva
 cy-preserving data mining were brought into existence. As is well-known\, 
 the offering of privacy\ncomes at a cost to data utility. Consequently\, a
 chieving a good balance between privacy and data utility is fundamental\nw
 hen designing privacy-preserving algorithms.\n\nIn this talk\, I will prov
 ide an overview of my recent research work in data anonymization and knowl
 edge hiding\, with\nemphasis on anonymization methods for medical data sha
 ring. These methods can increase data availability and\nutility\, which is
  important for supporting research on personalized medicine. This is becau
 se they provide privacy\nguarantees\, as well as they allow biomedical tas
 ks\, such as genome-wide association studies and clinical case\nanalysis\,
  to be performed accurately. To demonstrate the quality of the anonymizati
 on approaches\, I will present a\ncase study using data from the EMR syste
 m of the Vanderbilt University Medical Center (VUMC)\, a state-of-the-art\
 nsystem that stores information about 2 Million patients over 15 years.\n
LOCATION:Lecture Theatre 1\, Computer Laboratory
END:VEVENT
END:VCALENDAR