BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Detecting Sybil attacks and recommending social contacts from prox
 imity records - Daniele Quercia (University of Cambridge)
DTSTART:20100304T160000Z
DTEND:20100304T170000Z
UID:TALK22985@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:I’ll present two algorithms called MobID[1] and FriendSensin
 g[2]. Using short-range technologies (e.g.\, Bluetooth) on their mobile ph
 ones\, users keep track of other phones in their proximity. Upon proximity
  records\, MobID identifies Sybil attackers in a decentralized way\,  and 
 FriendSensing recommends social contacts:\n\n- The idea behind MobID is th
 at a device manages two small networks in which it stores information abou
 t the devices it meets: its network of friends contains honest devices\, a
 nd its network of foes contains suspicious devices. By reasoning on these 
 two networks\, the device is then able to determine whether an unknown ind
 ividual is carrying out a Sybil attack or not.\n\n- FriendSensing processe
 s proximity records using a variety of algorithms that are based on social
  network theories of geographical proximity and of link prediction. It the
 n returns a personalized and automatically generated list of people the us
 er may know.\nWe'll see how both algorithms perform against real mobility 
 and social network data.\n\n[1] Sybil Attacks Against Mobile Users: Friend
 s and Foes to the Rescue. Infocom '10\n\n[2] FriendSensing: Recommending F
 riends Using Mobile Phones. RecSys '09\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR