Hi , Please suggest me the Dataset for production quality translations. I have tried with Europarl-en-fr-(V7) data (~400MB) with Trigram-KenLM model i got BLEU score of 24.7.
But results are not satisfactory. My questions are, a) Which data set i should use for production quality translation or What should be size of dataset ? b) Do i need to consider other approaches over phrase based decoding such as *hierarchical or factored ?* I really appreciate if you can give me a head start. Thanks & Regards, Alind Billore -- Regards, Alind Billore
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
