BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Clinical De-Identification and Semantic Relatedness - Mohamed Abda
 lla\, University of Toronto
DTSTART:20211125T150000Z
DTEND:20211125T160000Z
UID:TALK166351@talks.cam.ac.uk
CONTACT:Marinela Parovic
DESCRIPTION:The first part of this talk will discuss the development of no
 vel low-cost approaches to de-identifying clinical notes. The second part 
 of the talk discuss the development of a new dataset of semantic relatedne
 ss for sentence pairs.. This dataset\, STR-2021\, has 5\,500 English sente
 nce pairs manually annotated for semantic relatedness using a comparative 
 annotation framework. We show that the resulting scores have high reliabil
 ity (repeat annotation correlation of 0.84). We use the dataset to explore
  a number of questions on what makes two sentences more semantically relat
 ed. We also evaluate a suite of sentence representation methods on their a
 bility to place pairs that are more related closer to each other in vector
  space.
LOCATION:https://cam-ac-uk.zoom.us/j/97599459216?pwd=QTRsOWZCOXRTREVnbTJBd
 XVpOXFvdz09
END:VEVENT
END:VCALENDAR