BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Kvasir: Scale up Latent Semantic Analysis-Based Content Provision 
 - Liang Wang
DTSTART:20150623T121500Z
DTEND:20150623T124500Z
UID:TALK59867@talks.cam.ac.uk
CONTACT:Heidi Howard
DESCRIPTION:The Internet is overloading its users with excessive informati
 on flows\, so that effective content-based filtering becomes crucial in im
 proving user experience and work efficiency. We build Kvasir\, a semantic 
 recommendation system\, on top of latent semantic analysis and other state
 -of-art technologies to seamlessly integrate an automated and proactive co
 ntent provision service into web browsing. \n\nThe presentation will focus
  on the architectural design of Kvasir\, and illustrate how to utilise dat
 a-parallel paradigm to scale up Kvasir into a practical Internet service. 
  In addition\, I will present the solutions to the technical challenges in
  the actual system implementation\, e.g\, improving the accuracy of classi
 c random-projection\, reducing index size and etc. In the last\, I will di
 scuss some promising directions for the future research.\n\nKvasir was dem
 onstrated in WWW’15. You can find the techreport\, demo videos and part 
 of the source code on the Kvasir website: http://www.cl.cam.ac.uk/~lw525/k
 vasir/
LOCATION:Computer Laboratory\, William Gates Building\, Room FW11
END:VEVENT
END:VCALENDAR