BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Inferring Interests from Mobility and Social Interactions - Anasta
 sios Noulas\, University of Cambridge
DTSTART:20091201T130000Z
DTEND:20091201T134500Z
UID:TALK21698@talks.cam.ac.uk
CONTACT:Stephen Kell
DESCRIPTION:In recent years there has been an explosion in the availabilit
 y of data sets about colocation between individuals and connectivity with 
 specific network infrastructure access points\, from which user location c
 an be inferred. These traces are usually collected through mobile devices 
 equipped with short-range radio inter- faces\, such as Bluetooth. Their po
 tential is enormous as user movement data can be mapped onto the geographi
 cal space and the social interactions of individuals can be extrapolated f
 rom the colocation data. Quite interestingly\, some of these data sets als
 o contain a description of user profiles\, such as the interests of the pe
 rson\, his/her age and gender and so on. In this paper we show that mobili
 ty and colocation information (i.e.\, social interactions) can be used to 
 infer user interests by applying standard machine learning techniques. We 
 evaluate a supervised and a semi-supervised technique using two different 
 data sets containing information of interactions amongst people at confere
 nces. We assume different degrees of available information for the inferen
 ce problem and show that we are able to predict people?s interests with go
 od accuracy also when only a small amount of information about user intere
 sts is available. While correlation of user interests with movement and pr
 oximity has already been investigated in social network research\, this is
  the first work that uses machine learning to show this quantitatively. 
LOCATION:Computer Laboratory\, William Gates Building\, Room FW11
END:VEVENT
END:VCALENDAR