BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Probabilistic anonymisation of microdatasets and models for analys
 is - Harvey Goldstein (University of Bristol\; University College London)
DTSTART:20160708T090000Z
DTEND:20160708T100000Z
UID:TALK66685@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:The general idea is to use the addition of random noise with k
 nown properties to some or all variables in a released dataset\, typically
  following linkage\, where the values of some identifier variables for ind
 ividuals of interest are also available to an external &lsquo\;attacker&rs
 quo\; who wishes to identify those individuals so that they can interrogat
 e their records in the dataset. The noise is tuned to achieve any given de
 gree of anonymity to avoid identification by an &lsquo\;attacker&rsquo\; v
 ia the linking of patterns based on the values of such variables.&nbsp\; T
 he noise so generated can then be &lsquo\;removed&rsquo\; at the analysis 
 stage since its characteristics are known\, requiring disclosure of these 
 characteristics by the linking agency. This leads to consistent parameter 
 estimates\, although a loss of efficiency will occur\, but the data themse
 lves are not degraded by any form of coarsening such as grouping.&nbsp\;
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR