<p>My french is rusty, but spanish would have a disambuation by prefixing the accustive with a preposition:<br>
E<br>
Hombre muerde a perro. (Articles omitted to correspond better with the example)</p>
<p>Mark</p>
<div class="gmail_quote">On Nov 20, 2011 7:16 PM, "Mike Maxwell" <<a href="mailto:maxwell@umiacs.umd.edu">maxwell@umiacs.umd.edu</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
In LILT 6 (<a href="http://elanguage.net/journals/index.php/lilt/issue/current" target="_blank">http://elanguage.net/<u></u>journals/index.php/lilt/issue/<u></u>current</a>), "Zipf's Law and l'Arbitraire du Signe," Martin Kay discusses statistical MT, and says (p.22):<br>
<br>
Notice that a language model would, and should, guarantee<br>
that the French “homme mord chien” would be translated into<br>
English as “dog bites man”, rather than “man bites dog”,<br>
which is what it really means.<br>
<br>
I once proposed this exact example (with Spanish rather than French) to a computational linguist who knew more about MT than I do. (People who know more about MT than I do are quite common. Ok, they're quite common among computational linguists :-).) That person suggested I needed to learn more about MT.<br>
<br>
It would be nice to find myself making the same mistake that Martin Kay made. It would be even nicer if it weren't a mistake.<br>
<br>
Is Kay's claim correct? The context is of course pure statistical MT, not hybrid rule/ statistical systems. Assume that the pair "homme mord chien"/ "man bites dog" never occurs in the training data, but that the reverse does (or at least that "dog bites man" appears on the English side, presumably with some significant frequency).<br>
-- <br>
Mike Maxwell<br>
<a href="mailto:maxwell@umiacs.umd.edu" target="_blank">maxwell@umiacs.umd.edu</a><br>
"My definition of an interesting universe is<br>
one that has the capacity to study itself."<br>
--Stephen Eastmond<br>
<br>
______________________________<u></u>_________________<br>
UNSUBSCRIBE from this page: <a href="http://mailman.uib.no/options/corpora" target="_blank">http://mailman.uib.no/options/<u></u>corpora</a><br>
Corpora mailing list<br>
<a href="mailto:Corpora@uib.no" target="_blank">Corpora@uib.no</a><br>
<a href="http://mailman.uib.no/listinfo/corpora" target="_blank">http://mailman.uib.no/<u></u>listinfo/corpora</a><br>
</blockquote></div>