BEGIN:VCALENDAR
VERSION:2.0
PRODID:ILLC Website
X-WR-TIMEZONE:Europe/Amsterdam
BEGIN:VTIMEZONE
TZID:Europe/Amsterdam
X-LIC-LOCATION:Europe/Amsterdam
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701025T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:/NewsandEvents/Archives/2007/newsitem/2126/12-
 December-2007-Computational-Linguistics-Seminar-Ph
 ilipp-Koehn
DTSTAMP:20071202T000000
SUMMARY:Computational Linguistics Seminar, Philipp
  Koehn
ATTENDEE;ROLE=Speaker:Philipp Koehn (Edinburgh)
DTSTART;TZID=Europe/Amsterdam:20071212T140000
DTEND;TZID=Europe/Amsterdam:20071212T000000
LOCATION:P.327, Euclides building, Plantage Muider
 gracht 24, Amsterdam
DESCRIPTION:Machine translation is more relevant t
 han ever, especially in a European Union with 23 o
 fficial languages. What will happen to languages s
 uch as Dutch? Will it survive as a language of com
 merce, or will it be abandoned in favour of Englis
 h? By lowering translation costs, we would expect 
 to systain the viability of a commercial zone that
  uses so many different languages. Statistical mac
 hine translation holds the promise of instant mach
 ine translation. Given open source tools such the 
 Moses decoder, just add a parallel corpus and you 
 have a machine translation system. This talk will 
 present some problems where the standard phrase-ba
 sed approach fails, and where attention to the spe
 cifics of the languages involved is required. I wi
 ll present methods that deal with different word o
 rder, morphology and agglutinative compounding.   
  For more information and abstracts, see https://w
 ww.illc.uva.nl/LaCo/CLS/
X-ALT-DESC;FMTTYPE=text/html:\n      <p>\n        
 Machine translation is more relevant than ever, es
 pecially\n        in a European Union with 23 offi
 cial languages. What will\n        happen to langu
 ages such as Dutch? Will it survive as\n        a 
 language of commerce, or will it be abandoned in f
 avour\n        of English? By lowering translation
  costs, we would expect\n        to systain the vi
 ability of a commercial zone that uses\n        so
  many different languages.\n        Statistical ma
 chine translation holds the promise of instant\n  
       machine translation. Given open source tools
  such the Moses\n        decoder, just add a paral
 lel corpus and you have a machine\n        transla
 tion system.\n        This talk will present some 
 problems where the standard\n        phrase-based 
 approach fails, and where attention to the\n      
   specifics of the languages involved is required.
  I will present methods that deal with different w
 ord order, morphology and agglutinative compoundin
 g.\n      </p>\n    \n      <p>\n        For more 
 information and abstracts,\n        see <a target=
 "_blank" href="https://www.illc.uva.nl/LaCo/CLS/">
 https://www.illc.uva.nl/LaCo/CLS/</a>\n      </p>\
 n    
URL:/NewsandEvents/Archives/2007/newsitem/2126/12-
 December-2007-Computational-Linguistics-Seminar-Ph
 ilipp-Koehn
END:VEVENT
END:VCALENDAR