Hi all, I write to you to grasp you attention on a paper by R.C. Carrasco and I in which we study different strategies to apply statistical machine translation techniques in order to retrieve documents that are a plausible translation of a given source document.
A piece of software implementing the different techniques we have tried can be freely (GPL license) downloaded from http://code.google.com/p/doctrans/. This software relies on Moses and may be a out-dated because of changes in Moses that made the direct access to the phrase table probabilities to fail; I hope to fix these things in the near future, meanwhile, you can use it with the revision of Moses I used for the experiments: rev. 2281. This software could be of interest for mining parallel corpora from comparable corpora. The paper is available at http://dx.doi.org/10.1080/08839514.2011.559906, if yo do not have access to the paper, please do not hesitate to contact me. Cheers -- Felipe Sánchez Martínez Dep. de Llenguatges i Sistemes Informàtics Universitat d'Alacant, E-03071 Alacant (Spain) Tel.: +34 965 903 400, ext: 2966 Fax: +34 965 909 326 http://www.dlsi.ua.es/~fsanchez _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
