BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BioNLP Research at Google - Ryan McDonald\, Google
DTSTART:20191114T110000Z
DTEND:20191114T120000Z
UID:TALK134767@talks.cam.ac.uk
CONTACT:Edoardo Maria Ponti
DESCRIPTION:In this talk I will highlight some ongoing efforts at Google o
 n improving discovery from biomedical literature. Topics will include rese
 arch in retrieval\, evidence extraction\, answer linking and network embed
 ding for biomedical ontologies. I will focus on how each of these interact
  to create a cohesive tool to accelerate literature search. Most of the ta
 lk will focus on document and evidence retrieval\, which is the entry poin
 t for all literature tools. I will discuss advances in deep learning in th
 is space and show they make a noticeable impact in quality for the domain.
  Specifically\, recent models that jointly extract relevant documents and 
 focused evidence passages can increase retrieval metrics substantially. Jo
 int work with many colleagues at Google Research and Athens University of 
 Economics and Business.
LOCATION:Board room\, Faculty of English\, 9 West Rd (Sidgwick Site)
END:VEVENT
END:VCALENDAR