RFERL corpora

From Apertium
Revision as of 18:41, 5 December 2013 by Firespeaker (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Radio Free Europe / Radio Liberty runs news services in a number of Central Asian languages. The information is essentially free for public use with attribution.

need link to usage info

We discovered how a corpus could be built from their website, and have instructions for writing a scraper in the framework we developed for it. Currently we have corpora for Kazakh and Kyrgyz, covering only a couple years' worth of articles.

Kyrgyz[edit]

2009[edit]

  • Number of stems: 4.1M
  • Coverage: ~87.4%

2010[edit]

  • Number of stems: 3.4M
  • Coverage: ~88%

Kazakh[edit]

2009[edit]

2010[edit]

  • Number of stems: 3.2M
  • Coverage: ~85.4%