BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Take-away TV: Recharging Work Commutes with Predictive Preloading 
 of Catch-up TV Content - King's College London
DTSTART:20160412T180000Z
DTEND:20160412T200000Z
UID:TALK65212@talks.cam.ac.uk
CONTACT:Dr Sobia Hamid
DESCRIPTION:Abstract: \nMobile data off-loading can greatly decrease the l
 oad on and usage ofcurrent and future cellular data networks by exploiting
  opportunistic and frequent access to Wi-Fi connectivity. Unfortunately\, 
 Wi-Fi access from mobile devices canbe difficult during typical work commu
 tes\, e.g.\, via trains or cars on highways. Inthis paper\, we propose a n
 ew approach: to preload the mobile device with contentthat a user might be
  interested in\, and thereby avoid the need for cellular data access.We de
 monstrate the feasibility of this approach by developing a supervised mach
 ine learning model that learns from user preferences for different types o
 f content\, and propensity to be guided by the UI of the player\, and pred
 ictively preload entire TV shows. Testing on a dataset of nearly 3.9 milli
 on sessions fromall over the UK to BBC TV shows\, we find that predictive 
 preloading can save over 71% of the mobile data for an average user.\n\nBi
 o: \nDmytro Karamshuk is a data scientists at King’s College London. His
  research interest is in machine learning approaches to modellingbehavior 
 of online users. He has previously worked onstudying watching patterns of 
 TV-streaming users (e.g.\,BBC iPlayer)\, mining geo-location social networ
 ks (e.g.\,Foursquare) and predicting user behaviour in imagebasedsocial pl
 atforms (e.g.\, Pinterest). Dmytro is a co-founder and a former CEO of Sta
 nfy.com. 
LOCATION:For venue details\, please register at www.meetup.com/Data-Insigh
 ts-Cambridge/
END:VEVENT
END:VCALENDAR