BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Probabilistic Record Linkage and Deduplication after Indexing\, Bl
 ocking\, and Filtering - Jared Murray (Carnegie Mellon University)
DTSTART:20160708T133000Z
DTEND:20160708T143000Z
UID:TALK66688@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:When linking two databases (or deduplicating a single database
 )&nbsp\;the number of possible links grows rapidly in the size of the data
 bases under consideration\, and in most applications it is necessary to fi
 rst reduce the number of record pairs that will be compared.&nbsp\;Spurred
  by practical considerations\, a range of indexing or blocking&nbsp\;metho
 ds have been developed for this task. However\, methods for inferring link
 age structure that account for indexing\, blocking\, and filtering steps h
 ave not seen commensurate development. I review the implications of indexi
 ng\, blocking and filtering\, focusing primarily&nbsp\;on&nbsp\;the popula
 r Fellegi-Sunter framework and proposing a new model to account for partic
 ular forms of indexing and filtering.&nbsp\;
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR