BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:How Can NLP Help Cure Cancer?  - Regina Barzilay Massachusetts Ins
 titute of Technology
DTSTART:20170810T120000Z
DTEND:20170810T130000Z
UID:TALK73841@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Majority of cancer research today takes place in biology and m
 edicine. Computer science plays a minor supporting role in this process if
  at all. In this talk\, I hope to convince you that NLP as a field has a c
 hance to play a significant role in this battle. Indeed\, free-form text r
 emains the primary means by which physicians record their observations and
  clinical findings. Unfortunately\, this rich source of textual informatio
 n is severely underutilized by predictive models in oncology. Current mode
 ls rely primarily only on structured data.\nIn the first part of my talk\,
  I will describe a number of tasks where NLP-based models can make a diffe
 rence in clinical practice. For example\, these include improving models o
 f disease progression\, preventing over-treatment\, and narrowing down to 
 the cure. This part of the talk draws on active collaborations with oncolo
 gists from Massachusetts General Hospital (MGH).\nIn the second part of th
 e talk\, I will push beyond standard tools\, introducing new functionaliti
 es and avoiding annotation-hungry training paradigms ill-suited for clinic
 al practice. In particular\, I will focus on interpretable neural models t
 hat provide rationales underlying their predictions\, and semi-supervised 
 methods for information extraction. \n
LOCATION:Auditorium\, Microsoft Research Ltd\, 21 Station Road\, Cambridge
 \, CB1 2FB
END:VEVENT
END:VCALENDAR