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/2003/newsitem/369/18-2
 2-August-2003-Adaptation-of-Automatic-Learning-Met
 hods-for-Analytical-and-Inflectional-Languages
DTSTAMP:20030206T000000
SUMMARY:Adaptation of Automatic Learning Methods f
 or Analytical and\n    Inflectional Languages
DTSTART;VALUE=DATE:20030818
DTEND;VALUE=DATE:20030822
LOCATION:Vienna, Austria
DESCRIPTION:Automatic (machine) learning approache
 s to any NLP task became a rich area with a variet
 y of methodologies. During the last years, its dev
 elopment made significant progress in the directio
 n of presenting new methods and, at the same time,
  their modifications. These modifications are of d
 ifferent nature and dependent on the language unde
 r consideration. The aim of the workshop is to pre
 sent and evaluate various modifications of the aut
 omatic learning methods originally developed for E
 nglish and declared as language independent. We ar
 e especially interested in automatic learning meth
 ods for the problems of morphological tagging and 
 parsing across languages with high level of inflec
 tion. Further, we encourage quantitative and quali
 tative comparison/evaluation studies across langua
 ges on the inputs and the outputs of the mentioned
  procedures. The workshop encourages reports of wo
 rk on:  Summarization of morphological and syntact
 ic features relevant for various automatic learnin
 g procedures.Tendencies of improvement of the auto
 matic learning methods. Presentation of implemente
 d modifications and their cross language evaluatio
 n.New/Latest algorithms for automatic learning.Hyb
 rid approaches (Although, there are trials to appl
 y hybrid approaches, it seems that the true key of
  how to combine the various parts has still not be
 en found and lies mainly in the success of analyzi
 ng the errors of each single component. Studies wh
 ich present the connection elements for a successf
 ul combination of diverse approaches are invited.)
  This workshop is part of ESSLLI'03. For more info
 rmation, see http://ckl.mff.cuni.cz/~alaf03/
X-ALT-DESC;FMTTYPE=text/html:\n      <p>\n        
 Automatic (machine) learning approaches to any NLP
  task\n        became a rich area with a variety o
 f methodologies.\n        During the last years, i
 ts development made significant\n        progress 
 in the direction of presenting new methods and,\n 
        at the same time, their modifications. Thes
 e modifications\n        are of different nature a
 nd dependent on the language under\n        consid
 eration. The aim of the workshop is to present and
 \n        evaluate various modifications of the au
 tomatic learning\n        methods originally devel
 oped for English and declared as\n        language
  independent. We are especially interested in\n   
      automatic learning methods for the problems o
 f morphological\n        tagging and parsing acros
 s languages with high level of\n        inflection
 . Further, we encourage quantitative and qualitati
 ve\n        comparison/evaluation studies across l
 anguages on the inputs\n        and the outputs of
  the mentioned procedures. The workshop\n        e
 ncourages reports of work on:\n      </p>\n      <
 ol>\n        <li>Summarization of morphological an
 d syntactic features\n        relevant for various
  automatic learning procedures.</li>\n        <li>
 Tendencies of improvement of the automatic learnin
 g methods.\n        Presentation of implemented mo
 difications and their cross\n        language eval
 uation.</li>\n        <li>New/Latest algorithms fo
 r automatic learning.</li>\n        <li>Hybrid app
 roaches (Although, there are trials to apply\n    
     hybrid approaches, it seems that the true key 
 of how to combine\n        the various parts has s
 till not been found and lies mainly in\n        th
 e success of analyzing the errors of each single c
 omponent.\n        Studies which present the conne
 ction elements for a successful\n        combinati
 on of diverse approaches are invited.)</li>\n     
  </ol>\n    \n      <p>This workshop is part of ES
 SLLI'03.\n        For more information, see <a tar
 get="_blank" href="http://ckl.mff.cuni.cz/~alaf03/
 ">http://ckl.mff.cuni.cz/~alaf03/</a>\n      </p>\
 n    
URL:/NewsandEvents/Archives/2003/newsitem/369/18-2
 2-August-2003-Adaptation-of-Automatic-Learning-Met
 hods-for-Analytical-and-Inflectional-Languages
END:VEVENT
END:VCALENDAR