MT-EQuAl

Description

MT-EQuAl (Girardi et al., 2014) is a toolkit for the human assessment of Machine Translation output.

MT-EQuAl provides an annotation interface to carry out three annotation tasks:

  • quality rating of translations (e.g. adequacy/fluency, relative ranking)
  • annotation of translation errors
  • word-alignment

MT-EQuAl is web-based and multi-user, allowing large scale and remotely managed manual annotation projects. It incorporates a number of project management functions and sophisticated progress monitoring capabilities.

MT-EQuAl supports the following browsers: Chrome, Safari, Firefox, and Internet Explorer.

Acknowledgment

The development of MT-EQuAl was partially supported by the EU-funded project MateCat (ICT-2011.4.2-287688).

License

MT-EQuAl is distributed under the Apache License, Version 2.0.

Manuals

Installation and configuration instructions are available here.

Usage instructions for project management are available here.

Usage instructions for annotation can be found here.

Source code

The source code is available here.

Reference

If you intend to use MT-EQuAl please cite:

Girardi, Christian, Luisa Bentivogli, Mohammad Amin Farajian and Marcello Federico. 2014, “MT-EQuAl: a Toolkit for Manual Assessment of Machine Translation Output”. Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: System Demonstrations. Dublin, Ireland, pp. 120–123. (pdf)

Contacts

For questions and support about MT-EQuAl please contact: cgirardi [at] fbk [dot] eu