BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Detecting Temporal Sybil Attacks - Neal Lathia (UCL)
DTSTART:20100429T150000Z
DTEND:20100429T160000Z
UID:TALK23674@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:Recommender systems are vulnerable to attack: malicious users 
 may deploy a set of sybils to inject ratings in order to damage or modify 
 the output of Collaborative Filtering (CF) algorithms. Previous work in th
 e area focuses on designing sybil profile classification\nalgorithms: to p
 rotect against attacks\, the aim is to find and isolate any sybils. These 
 methods\, however\, assume that the full sybil profiles have already been 
 input to the system. Deployed recommender systems\, on the other hand\, op
 erate over time: recommendations may be damaged as sybils inject profiles 
 (rather than only when all the malicious ratings have been input)\, and sy
 stem administrators may not know when their system is under attack. In thi
 s work\, we address the problem of temporal sybil attacks\, and propose an
 d evaluate methods for monitoring global\, user and item behaviour over ti
 me in order to detect rating anomalies that reflect an ongoing attack. We 
 conclude by discussing the consequences of our temporal defenses\, and how
  attackers may design ramp-up attacks in order to circumvent them.\n\nBio:
  Neal is a Research Fellow in the Department of Computer Science\, Univers
 ity College London\, working on the EU iTour project with Dr Capra (http:/
 /www.itourproject.com). His PhD thesis (to be imminently\nsubmitted) was s
 upervised by Prof. Hailes and titled "Evaluating Collaborative Filtering O
 ver Time\;" the thesis dealt with modeling\, evaluating\, and improving th
 e temporal performance of recommender systems. More details are available 
 on: \nhttp://www.cs.ucl.ac.uk/staff/n.lathia\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR