BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:SwiftKey Keyboard: Language Modelling and Beyond - Caroline Gasper
 in\, Language Team Lead\, SwiftKey
DTSTART:20160128T130000Z
DTEND:20160128T140000Z
UID:TALK62376@talks.cam.ac.uk
CONTACT:Helen Yannakoudakis
DESCRIPTION:SwiftKey is the leading technology company developing smart ke
 yboard apps for mobile devices. SwiftKey applies AI to the problem of typi
 ng on touchscreens\, learning from each user to accurately autocorrect and
  predict their next words - the company's software features on more than 2
 50m devices worldwide and supports more than 100 languages. This talk focu
 ses on key challenges of mobile text input\, in particular multilingual la
 nguage modelling for text prediction. I'll describe SwiftKey's core techno
 logy and touch on our latest developments for improving our predictive cap
 acity.\n\n*Short bio:*\nCaroline Gasperin is the lead of SwiftKey's Langua
 ge team\, which is responsible for building the language models used by Sw
 iftKey's language inference engine. Before joining SwiftKey Caroline was a
 n academic researcher working on Natural Language Processing\, more specif
 ically on automatic text simplification\, text readability assessment\, an
 d anaphora resolution. Caroline received her PhD at the Computer Lab in 20
 08\, she was part of the NLIP group.
LOCATION:Computer Laboratory\, William Gates Building\, Room FW26
END:VEVENT
END:VCALENDAR