BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Linguistic Indicators for Estimating the Quality of Machine Transl
 ations - Mariano Felice\, University of Cambridge
DTSTART:20130215T120000Z
DTEND:20130215T130000Z
UID:TALK43216@talks.cam.ac.uk
CONTACT:Ekaterina Kochmar
DESCRIPTION:This talk presents a summary of research done as part of my\nm
 aster's thesis\, where I have addressed the problem of estimating the\nqua
 lity of machine translations automatically without having access to\nhuman
  references. Estimations are obtained by building a supervised\nregression
  model that predicts quality scores using features from the\nsource and ta
 rget texts as well as additional resources. Although shallow\nindicators a
 re generally used to characterise the relationship between the\nsource tex
 t and its translation\, they convey no notion of meaning\, grammar\nor lin
 guistic correctness so final estimations may be very biased towards\nsuper
 ficial aspects. The work presented here attempts to bridge that gap by\nin
 troducing more linguistic features and analysing how they compare with\nsh
 allow indicators over a set of publicly available datasets.
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR