BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:High-Dimensional Incremental Divisive Clustering u
 nder Population Drift - Pavlidis\, N (Lancaster Un
 iversity)
DTSTART;TZID=Europe/London:20140115T153000
DTEND;TZID=Europe/London:20140115T160000
UID:TALK49921AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/49921
DESCRIPTION:Clustering is a central problem in data mining and
  statistical pattern recognition with a long and r
 ich history. The advent of Big Data has introduced
  important challenges to existing clustering metho
 ds in the form of high-dimensional\, high-frequenc
 y\, time-varying streams of data. Up-to-date resea
 rch on Big Data clustering has been almost exclusi
 vely focused on addressing individual aspects of t
 he problem in isolation\, largely ignoring whether
  and how the proposed methods can be extended to a
 ddress the overall problem. We will discuss an inc
 remental divisive clustering approach for high-dim
 ensional data that has storage requirements that a
 re low and more importantly independent of the str
 eam size\, and can identify changes in the populat
 ion distribution that require a revision of the cl
 ustering result.\n
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:Mustapha Amrani
END:VEVENT
END:VCALENDAR