mkarasek edited this page Mar 6, 2013 · 15 revisions
Clone this wiki locally

M4Loc processing of ITS2.0 enhanced XLIFF files

This repository is dedicated to the library, web-service and test site for M4Loc process applied on round-tripping of XLIFF files enhanced by ITS 2.0 metadata.

The mlwlt-m4loc-xliff-mt service provides the detection of encoded ITS 2.0 metadata within XLIFF file and preparation for the Machine translation (MT) using M4Loc process. Supported ITS2.0 data categories are:

  • Translate - Translation units or sub-segments comming to XLIFF with ITS Translate="no" attribute are omitted from the MT translation process.
  • Domain - Algorithm of MT engine selection is based on ITS Domain metadata. Implemented for Moses MT engines using M4Loc preparation process.
  • Text Analysis - Implemented mechanism of translation sub-segments with Text Analysis data category defined. Such sub-segments are translated using resources defined in TA metadata instead of MT.
  • MT Confidence - Value from 0 to 1 describing the quality of MT output for given segment (applied on level.
  • Provenance - Information about data provenance.

The process of original M4Loc is modified, documentation and source code for original M4Loc can be found here.

This work is a part of ITS 2.0 implementations in Multilingual-Web LT working group.

Multilingual-Web LT