BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY: Processing Multiword Expressions for Grammatical Error Correction
  - Shiva Taslimipoor (University of Cambridge)
DTSTART:20230203T120000Z
DTEND:20230203T130000Z
UID:TALK196492@talks.cam.ac.uk
CONTACT:Rami Aly
DESCRIPTION:Abstract:\n\nMultiword expressions (MWEs) are combinations of 
 two or more words with syntactic and semantic idiosyncratic behaviours. Th
 ey are prevalent in any language and domain. Different categories of MWEs 
 include idioms\, nominal compounds\, light verb constructions\, verb parti
 cle constructions\, and each pose particular challenges in processing. Whi
 le they are known to be processed faster by native speakers\, language lea
 rners find them difficult to understand and use. Like most machine transla
 tion (MT) systems\, current Grammatical Error Correction (GEC) systems do 
 not take them into consideration and are not good at correcting them. \nIn
  this talk\, I give a brief presentation of a survey on different approach
 es to model MWEs in NLP. I summarise on how transformers advanced in deali
 ng with them and what aspects are still lacking proper solutions. And fina
 lly\, I illustrate our method for processing MWEs in a grammatical error c
 orrection system able to capture this type of errors better than baseline 
 GEC systems.\n\n\nBio:\n\nShiva Taslimipoor is a postdoctoral research ass
 ociate in the NLIP group at the University of Cambridge. She is a member o
 f ALTA Institute where her research focus is on developing tools for autom
 atic language learning and assessment. Before that she was a research asso
 ciate at RGCL at the University of Wolverhampton where she completed her P
 hD in the area of Natural Language Processing on the topic of ‘Automatic
  Identification and Translation of Multiword Expressions’.
LOCATION:Computer Laboratory\, Room FW09
END:VEVENT
END:VCALENDAR