>
> Have you done any evaluation ? How does it compare to other systems (and
> the old system too) ? :)
>
The pair works fairly well with encyclopedia-like texts, and has a good
Wikipedia coverage (92% for English and 87% for Catalan). The reference
translation (an English article on Greece not used during development)
shows a WER/PER of 51%/35%, better than the old pair's 56%/40% with the
same text. Yandex is slightly better than Apertium, with 56%/34%, and
Google stands with the best results (43%/26%). I have not really evaluated
translations from Catalan (most of the development has taken place in the
other direction), but it should be more or less the same as the old pair.
While the pair still needs a lot of work and love, the rewrite has eased
development. With good taggers on both sides, trained with diverse texts
(including dialogues to reflect oral language constructions), as well as a
reorganization/rewrite of the transfer rules (inherited from the messy old
pair), we should have a very decent and useful language pair.
Thanks for your support!
Marc
2018-03-11 23:59 GMT+01:00 Francis Tyers <fty...@prompsit.com>:
> El 2018-03-11 23:41, Marc Riera Irigoyen escribió:
>
>> Dear Apertiumers,
>>
>> After intense work during last year's GSoC and the following months,
>> I'm glad to announce that the apertium-eng-cat pair, currently in
>> apertium-incubator, is finally testvoc clean and ready for trunk. This
>> is a rewrite and a replacement of the original English-Catalan (en-ca)
>> pair, which was becoming increasingly out of date and hard to
>> maintain.
>>
>> The new pair includes everything the old pair did (rule-wise), but has
>> a considerably larger dix (~65,000 stems) and features several
>> innovations compared to what we had before:
>>
>> * Lexical selection rules (mainly eng>cat)
>> * Perceptron tagger for English
>> * Constraint Grammar
>> * Apertium-separable module
>>
>> These changes are important not only because of the improvements, but
>> also because Java compatibility cannot be kept (as with
>> apertium-fra-cat). As there is no possible fallback mode in the new
>> pair until these modules get ported to Java or a different approach
>> with C++/Java is taken, the best idea could be to temporarily use the
>> old pair as fallback for the new one.
>>
>> I will keep doing my best to improve the quality of this pair and use
>> it as a test bench for innovative modules and development approaches.
>>
>>
> PS. Congratulations on the release! :)
>
> F.
>
--
*Marc Riera Irigoyen*
Freelance Translator EN/JA>CA/ES
(+34) 652 492 008 <+34%20652%2049%2020%2008>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Apertium-stuff mailing list
Apertium-stuff@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
