Contact

If you have any trouble using DeepQuest, please open a GitHub issue, or drop an email to Lucia Specia.

License

DeepQuest is available under the BSD license. For pre-existing code and resources, e.g., scikit-learn, SRILM, GIZA++, Stanford and Berkeley parsers, please check their website.

Acknowledgement

The original work on deepQuest has been funded by the EAMT.

The implementation of the deepQuest framework follows the architecture and style of the NMT-Keras library, developed by Marc Bolaños and Álvaro Peris.

Related projects

Lucia Specia is a pioneer in Quality Estimation, and with her team, she has released numerous resources to support research work on this topic:

• Quest++: open source toolkit for pipelined multi-level translation Quality Estimation for statistical Machine Translation, based on feature extraction (written in Java)

• Marmot: open source toolkit for word-level QE (written in Python)

• Resources: a collection of datasets to support work on Quality Estimation

How to contribute?

If you wish to support the development of DeepQuest, by contributing to its code, or by making your approach available through it, please get in touch by sending an email to Lucia Specia.

How to cite DeepQuest

If you are using DeepQuest for your work, please cite it as follows:

DeepQuest: a Framework for neural-based Quality Estimation – Julia Ive, Frédéric Blain, Lucia Specia. In the Proceedings of COLING 2018, the 27th International Conference on Computational Linguistics, Sante Fe, New Mexico, USA, (2018). (PDF)