[Moses-support] Questions about the LGPL license for Moses

Hi, I just what to confirm what the LGPL license for Moses means: If some of us modify its codes, or develop a new MT engine based on Moses, can we use this new MT engine for commercial use (distribute it as a commercial product)? Or at least use it internally in our company? Thanks so much, Wen

Re: [Moses-support] Train Moses Engine for EN to ZH_CN

Hi Barry, Thanks for your information. I am still not sure about what the 'tokenizing' and the 'detokenizing' is, I mean, what they did and why those handlings are needed. Is the 'tokenizing' something the same with the segmenting? BTW, I am not familar with Java. Is there any such script wrote

Re: [Moses-support] Train Moses Engine for EN to ZH_CN

Hi Francois, Thanks for your information. Can you help to let me know or send your scripts for the Chinese tokenizing and detokenizing for my information? It seems the default tokenizing script doesn't support the Chinese language code. Thanks so, Wenlong 2010/8/31 > Send Moses-support mail

[Moses-support] Train Moses Engine for EN to ZH_CN

Hi, Is there anybody can share your experience about how to train a Moses engine from Engish to Simplified Chinese? I know there are some differences and some the training/tuning scripts are not supporting this language pair as default. I intend to train one Moses engine for this langauge pair, c

[Moses-support] Is there anyway to run the training steps one bt one?

Hi all, >From the training manual, we can see that, there are totally 9 steps for a new engine's training: Steps: (--first-step to --last-step) (1) prepare corpus (2) run GIZA (3) align words (4) learn lexical translation (5) extract phrases (6) score phrases (7) learn reordering model (8) learn g

[Moses-support] About Moses Web server

Hi All, We made a web service based on the information on this page: http://www.statmt.org/moses/?n=Moses.WebTranslation The web service works, but I can see some flaws in it: It only calls the Moses engine itself to do the translation in the daemon.plscript, but during our evaluation steps, the

Re: [Moses-support] What will happen if I used a wrong language model during the Moses training?

into translation without a language model. > > I'd suggest simply to train a correct language model and repeat the > MERT and evaluation steps, unless you are interested in your setup as > a researcher :) > > Best, > Mark > > On Tue, Aug 10, 2010 at 5:04 PM, Wen

[Moses-support] What will happen if I used a wrong language model during the Moses training?

Hi, I have trained a Moses Engine for the language pair English to Spanish, but for the training script, during below step: "nohup nice tools/moses-scripts/scripts-20100203-0505/training/train-factored-phrase-model.perl -scripts-root-dir tools/moses-scripts/scripts-20100203-0505 -root-dir Work-SP

Re: [Moses-support] For the BLEU/NIST evaluation

idence intervals with bootstrap resampling, which > will give you some indication how reliable a, say, 0.5, 1.0, or 2.0 point > difference in BLEU is. > > Regarding test set sizes, it should be at least 1000 sentence pairs, > 12,000 is certainly very large. > > -phi > > On S

Re: [Moses-support] For the BLEU/NIST evaluation

, Wenlong Yang : > Hi all, > > can any of you help to provide some materials about how to select the > sample > data for BLEU/NIST evaluation? > I mean, how many lines of data shoud I choose for the evaluation? and how > can I choose the data to let them can be more representable

[Moses-support] How to use relevant training data to train a good engine for our domain?

Hi Guys, I have a question here: I want to train a moses engine for domain A, now I have some training data for domain A (for example, 4 words) and more training data (for example, 20 words) which is not specifically belongs domain A, but also relevant. How can I use the extra training da

[Moses-support] For the BLEU/NIST evaluation

Hi all, can any of you help to provide some materials about how to select the sample data for BLEU/NIST evaluation? I mean, how many lines of data shoud I choose for the evaluation? and how can I choose the data to let them can be more representable for our domain/use? I have tried to generate B

Re: [Moses-support] Questions about Moses: terminology database

ls > http://www.statmt.org/moses/?n=Moses.AdvancedFeatures#ntoc4 > > best regards > Barry > > On Tuesday 03 August 2010 07:48, Wenlong Yang wrote: > > Hi, > > > > I have set up the Moses, but when I used it to do the machine > transaltion, > > I can not find whet

[Moses-support] Questions about Moses: terminology database

Hi, I have set up the Moses, but when I used it to do the machine transaltion, I can not find whether there is somewhere I can set the dictionary (terminoloy database) for it. Does anyone know how to set this in Moses? Or shall we need to modify the Moses's code? Does anybody has the same experie

[Moses-support] low BLEU score for Moses engine

Hello, We have setup the Moses engine on my desktop, and have run the pilot training from EN to FR by following the steps listed at http://www.statmt.org/moses_steps.html and got the BLEU score as 25.20. Then we use more training data (maybe 343431 segments) for a specific technical doamin to tra