BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Monday Seminar :" Watch My Behavior\, Find Her Content"
DTSTART:20110307T161500
DTSTAMP:20261005T111042Z
UID:bbe9920b7b2faeceb98cb4c7b00da898141f9e628b2002c1a82c9264
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Sihem Amer-Yahia\, Senior Research Scientist at Yahoo! Res
 earch\, Barcelona - IC Faculty candidate\nAbstract: On the Social Web\, us
 er behavior is central to defining content relevance and to finding conten
 t efficiently. The behavior of each information seeker identifies a unique
  subset of the population\, the seeker's network\, whose opinion matters. 
 While search over information popular in a seeker's network can be achieve
 d efficiently with straightforward adaptations of well-known algorithms\, 
 the storage volume of indexing every (seeker\, keyword) pair\, is daunting
 . As a result\, I will investigate two space-saving solutions\, network cl
 ustering and behavior clustering. A similar concern arises in group recomm
 endation. Content relevance to a group is a function of members' disagreem
 ent\, and efficient recommendation requires maintaining pair-wise user dis
 agreement indexes\, another storage pit. Hence\, I will explore two space-
 saving strategies for group recommendation\, behavior factoring and partia
 l materialization. I will argue that scalable and efficient network-aware 
 search and group recommendation must leverage user behavior to balance sto
 rage volume and response time. Finally\, I will discuss my future work\, t
 hat is\, the need to structure the crowds toward achieving a common goal a
 nd the technical challenges raised by building a collaborative platform to
  do so.\n\nBio: Sihem spent 7 years as Member of Technical Staff at AT&T L
 abs and has been Senior Research Scientist at Yahoo! since May 2006. She h
 olds a Ph.D. in Computer Science from U. Paris-Orsay and INRIA. Her resear
 ch is on data management\, query processing and relevance models that leve
 rage social behavior for online content serving. Sihem has chaired SIGMOD 
 2009 Tutorials and several conference tracks including CIKM 2008\, VLDB 20
 09\, ICDE 2010\, WWW 2010\, SIGMOD 2011\, and EDBT/ICDT 2012. She is a mem
 ber of the Board of Trustees of the VLDB Endowment and of the ACM SIGMOD E
 xecutive Committee. Sihem serves as the Area Editor for TODS\, VLDB Journa
 l\, and Information Systems
LOCATION:INM 202
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR
