BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Recommending Investors for Crowdfunding Projects - Jisun An (Unive
 rsity of Cambridge)
DTSTART:20140313T150000Z
DTEND:20140313T160000Z
UID:TALK50053@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:To bring their innovative ideas to market\, those embarking in
  new ventures have to raise money\, and\, to do so\, they have often resor
 ted to banks and venture capitalists. Nowadays\, they have an additional o
 ption: that of crowdfunding. The name refers to the idea that funds come f
 rom a network of people on the Internet who are passionate about supportin
 g others’ projects. One of the most popular crowdfunding sites is Kickst
 arter. In it\, creators post descriptions of their projects and advertise 
 them on social media sites (mainly Twitter)\, while investors look for pro
 jects to support. The most common reason for project failure is the inabil
 ity of founders to connect with a sufficient number of investors\, and tha
 t is mainly because hitherto there has not been any automatic way of match
 ing creators and investors. We thus set out to propose different ways of r
 ecommending investors found on Twitter for specific Kickstarter projects. 
 We do so by conducting hypothesis-driven analyses of pledging behavior and
  translate the corresponding findings into different recommendation strate
 gies. The best strategy achieves\, on average\, 84% of accuracy in predict
 ing a list of potential investors’ Twitter accounts for any given projec
 t. Our findings also produced key insights about the whys and wherefores o
 f investors deciding to support innovative efforts.
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR