Dear Apertiumers,
I'm happy to announce the first beta release (0.1.0) of the arg-cat (Aragonese
and Catalan) translator, using the arg and cat monolingual modules. Fran Tyers
and I have been working on it during the last couple of weeks.
The main features are presented below.
* Initial release (0.1.0), beta
* Operative bidirectional release.
* Features:
* Uses monolingual packages apertium-cat and apertium-arg (separated
monolingual and bilingual data).
* Trimming of the monolingual dictionary (in both directions).
* Lexical selection support (not used yet).
* Monolingual dictionaries:
Aragonese (apertium-arg): > 546 paradigms, > 21073 lemmae (of which,
8047 proper nouns), including 1921 multi-words.
Catalan (apertium-cat): > 559 paradigms, > 41143 lemmae (of which, 8341
proper nouns), including 4115 multi-words.
* Bilingual dictionary: 24110 entries. Open categories crossed from
apertium-spa-arg (ver 0.4.1) and apertium-es-ca. Some closed categories
incomplete.
* Naïve coverage:
Aragonese: From 87.7% in an.wikipedia to 89.2% in a corpus of narrative
texts.
Catalan: From 87.6% in Catalan corpus at
trunk/apertium-eo-ca/tekstaro/ca.crp.txt to 93.2% at
trunk/apertium-es-ca/ca-tagger-data/ca.tagged.txt
* Translation performance: Catalan to Aragonese (50 first sentences of
ca.crp.txt): WER = 19.37% (PER=17.85%); WER (unknown removed) = 15.48%
(PER=13.96%). Unknown words: 11.86%. Free rides: 32.69%.
* Dialectal support: Analysis of Aragonese dialectal forms supported.
Valencian forms are analyzed, but not generated.
* Main pending Issues:
* Improve on coverage.
* Work on transfer.
* Develop lexical selection rules
Best,
Juan Pablo
<http://sourceforge.net/about>
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
_______________________________________________
Apertium-stuff mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
