Search results

• Apertium-nog/stats
  == Corpora ==
  1 KB (176 words) - 06:05, 16 December 2014
• Apertium-ukr/stats
  == Corpora ==
  1 KB (158 words) - 05:29, 22 August 2017
• Aromanian
  ===Corpora===
  8 KB (1,048 words) - 05:32, 1 December 2017
• Kashmiri
  == Other Corpora ==
  6 KB (811 words) - 10:42, 2 July 2018
• Apertium-uzb/stats
  == Corpora ==
  3 KB (324 words) - 21:41, 15 December 2019
• Apertium-sqi/stats
  == Corpora ==
  1 KB (154 words) - 06:16, 16 December 2014
• Apertium-fin/stats
  == Corpora ==
  328 bytes (35 words) - 05:49, 16 December 2014
• English and Catalan/Workplan
  * Tagger training preparation (tagged corpora unification) * Tagger training preparation (tagged corpora unification)
  5 KB (506 words) - 14:56, 28 August 2017
• Apertium-kum/stats
  == Corpora ==
  2 KB (235 words) - 05:58, 16 December 2014
• Apertium Turkic
  ...Pirinen, Jonathan Washington (2015). "Finite-state morphologies and text corpora as resources for improving morphological descriptions". [https://sites.goog
  13 KB (1,710 words) - 20:32, 30 August 2018
• Unsupervised tagger training
  # the best taggers use hand-tagged corpora to train with (we use untagged corpora -- for English)
  7 KB (1,177 words) - 08:34, 8 October 2014
• Traductions en français
  * [[Corpora]]
  13 KB (1,601 words) - 23:31, 23 July 2021
• Lextor
  to be related. Both corpora must be pre-processed before the training. This pre-processing, consisting in analysing the corpora and
  11 KB (1,814 words) - 03:22, 9 March 2019
• Helsinki Apertium Workshop/Session 8
  The final coverage of the system was around 90%, e.g. over a set of corpora 10 unknown words out of 100 on average. The word-error rate was around 17%, When linguistic resources, for example corpora, dictionaries, grammars, morphological analysers, lists of lemmata etc. are
  12 KB (1,683 words) - 08:42, 10 May 2013
• Tartu Apertium Course/Session 8
  The final coverage of the system was around 90%, e.g. over a set of corpora 10 unknown words out of 100 on average. The word-error rate was around 17%, When linguistic resources, for example corpora, dictionaries, grammars, morphological analysers, lists of lemmata etc. are
  12 KB (1,683 words) - 11:00, 30 October 2015
• Automatically generating compound bidix entries
  ...toscore.txt | ~/source/apertium/trunk/apertium-lex-learner/irstlm-ranker ~/corpora/català/en.blm > unk.trans.scored.txt
  9 KB (1,470 words) - 11:28, 24 March 2012
• Курсы машинного перевода для языков России/Session 8
  The final coverage of the system was around 90%, e.g. over a set of corpora 10 unknown words out of 100 on average. The word-error rate was around 17%, When linguistic resources, for example corpora, dictionaries, grammars, morphological analysers, lists of lemmata etc. are
  12 KB (1,679 words) - 12:00, 31 January 2012
• Turkic-Turkic translator
  * '''Coverage''' is the naïve coverage over one or more free corpora.
  6 KB (591 words) - 22:50, 30 October 2017
• Quick and dirty guide addendum: other important things
  ...steps to create each one. But you'll want to build it up, testing against corpora, etc. You want to be able to have as many correct analyses as possible, an ...if your pair will support translation in both directions). The corpus or corpora ideally should represent a range of content—i.e., it shouldn't be just sp
  10 KB (1,615 words) - 07:43, 20 December 2015
• Apertium cat-srd and ita-srd/GSoC 2017
  ...dinian to Italian. We started a manual morphological disambiguation of the corpora that will help the translator to recognize the correct morphology of each w We have treated two corpora: one journalistic and more dialectal, and other taken directly from literar
  9 KB (1,306 words) - 15:56, 2 September 2017

View (previous 20 | next 20) (20 | 50 | 100 | 250 | 500)