Waxp License Code

WAT allows to align parallel texts using various aligning engines:

• JAM : Written in Perl. Allows to align more than two languages at the same time.
• Alinea Lite : Written in Prolog, a long time ago. Slow but robust.
• LF Aligner : a project leaded by Andras Farkas, based on Daniel Varga's Hunalign.
• YASA : Originally created by Philippe Langlais, it was reimplemented by Alexandre Patry. It was reborn as yasa as part of improvements made during Fethi Lamraoui's masters thesis.

References

Kraif O. (2001) Exploitation des cognats dans les systèmes d’alignement bi-textuel : architecture et évaluation, TAL 42 :3, ATALA, Paris, pp. 833-867. (pdf)

Kraif Olivier (2015) Multi-alignement vs bi-alignement : à plusieurs, c’est mieux ! , Actes de TALN 2015, 22ème Conférence sur le Traitement Automatique des Langues Naturelles, Caen, 22-25 juin 2015, pp. 255-266 (pdf)

Lamraoui, F., and P. Langlais (2013) "Yet Another Fast, Robust and Open Source Sentence Aligner. Time to Reconsider Sentence Alignment?", XIV Machine Translation Summit, Nice, France. (pdf)

Varga, D., Németh, L., Halácsy, P., Kornai, A., Trón, V., Nagy, V. (2005). Parallel corpora for medium density languages. In Proceedings of the RANLP 2005, pages 590-596. (pdf)

Web design

Built with the Bootstrap CSS framework and Icons from Glyphicons.

File Upload Widget imported from jQuery-File-Upload.

File Tree Widget imported from jQueryFileTree.

Language identification

Automatic language identification uses guessLanguage.js library

Some explanations about the file naming scheme...

When you upload various files on the site, you have to follow a strict naming scheme, in order to allow the system to recognize what files are corresponding with each other, and what are the involved languages. By default, the naming scheme used by the system uses the following conventions : NAME.LL.EXT :

• NAME is the common substring shared by all the files that correspond to a same text
• LL is the language suffix (for instance 'fr' for French, 'en' for English , 'de' for German, etc. You can use your own codes, for instance 'f1','f2','f3' if you align three French versions of a same text.
• EXT indicates which format is used (txt=raw text, ces=cesAna text, xml=xml format, etc.)

For instance : balzac.fr.txt, balzac.en.txt

If you use a different naming scheme, you will have to redefine some Perl regular expressions in advanced parameters :

• File name pattern: this pattern describes which files have to be processed. The first capturing group (first parentheses) indicates the common name string.
• Language pattern: when using YASA or Alinea, in complement to file name pattern, this regex pattern describes where is defined the language in the file names. The capturing group defines the language code in the name. By default, the language string has two characters, but it may be changed.
• Aligning file name: this template indicates how Yasa or Alinea should name the aligned file. The default value is : '.-.'

How are structured output directories ?

In the output files tab, you will find all the uploaded files in the root directory. Every time you run an aligning session, a new output_* directory is created : this directory contains all the output files.

• By default, when using LF Aligner, output dir contains TMX and CSV formats
• When using YASA, output dir contains the various bilingual YASA outputs in CES format, and a TMX version. If you align more than two languages, the transitively merged alignment is produced only for TMX.
• When using Jam, output dir either the bilingual and multi-alignment in TMX format.
• When using Alinea lite, output dir contains only a bilingual output in TMX format.