I have two questions about Moses: (1) I realized that the symmetrization heuristics do not produce the same result running Moses in different language directions. In particular, the model/lex.* files are slightly different (different pairs and also different scores). I expected the lex.e2f to be the same as the lex.f2e in the other direction but it's not. (but maybe that's normal). This should also effect the phrase-extraction, shouldn't it? (I didn't check). Are there any consequences?
(2) about factored models: is it possible to train generation steps from other (monolingual) corpora (not the bitext) with Moses? Or would I have to implement this myself? Should I expect a (significant) gain like when using larger LM's? Anyone who tried that already? Thanks in advance for any thoughts, -- Jörg ***********/\/\/\/\/\/\/\/\/\/\/\************************************ ** Jörg Tiedemann [EMAIL PROTECTED] ** ** Alfa-Informatica http://www.let.rug.nl/~tiedeman ** ** Rijksuniversiteit Groningen Harmoniegebouw, room 1311-429 ** ** Postbus 716 phone: +31 (0)50-363 5935 ** ** 9700 AS Groningen fax: +31 (0)50-363 6855 ** *************************************/\/\/\/\/\/\/\/\/\/\/\********** _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
