BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Towards Real-Time Crowd-Oriented Search and Computation
DTSTART:20111028T110000
DTSTAMP:20260406T214446Z
UID:56b4e3b5fa13cb389041a82a331bc148d29b98801f77458fbde0fe17
CATEGORIES:Conferences - Seminars
DESCRIPTION:James Caverlee\nOnline communities are the fastest growing phe
 nomena on the Web\, enabling millions of users to discover and explore com
 munity-based knowledge spaces. While long-lived communities have been one 
 of the key organizing principles of these systems\, there is widespread ev
 idence of highly-dynamic\, ad-hoc crowd formation in existing social syste
 ms. Examples range from users posting to Facebook in response to a live Pr
 esidential debate\, to users sharing pictures about a chemical fire at a n
 earby refinery\, to blog commenting about breaking news\, and so on. These
  crowds are dynamically formed and potentially short-lived\, often with on
 ly implicit signals of their formation at all. Identifying these highly-dy
 namic crowds from the massive scale of the real-time web\, monitoring thei
 r quality\, and connecting stakeholders to these crowds in real-time could
  revolutionize the decision-making of critical stakeholders. Unfortunately
 \, existing search solutions cannot be directly applied to nascent crowds\
 , leaving a significant research gap. In this talk\, I will highlight our 
 work towards developing real-time crowd-oriented search and computation sy
 stems\, so that high-value stakeholders can monitor\, analyze\, and distil
 l high-quality information from bursty social systems and actively engage 
 with the crowds generating this information.
LOCATION:INR219
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR