|
Welcome to the website of the Machine Translation (MT) group in the School of Computing at DCU. The research
that we carry out is undertaken in the National Centre for Language Technology (NCLT), and the Centre for Next Generation Localisation (CNGL).
We carry out a wide range of research in MT in the NCLT and CNGL, including: |
- Integrating Syntax & Semantics into Phrase-Based Statistical Machine Translation
- Hybrid Statistical & Example-Based Machine Translation
- Tree-Based Machine Translation
- Sub-sentential Alignment for Machine Translation
- Spoken Language Translation
- Probabilistic Transfer-Based Machine Translation
- Evaluation in Machine Translation
- Controlled Language & Machine Translation
- Human Factors in Machine Translation
- Quality Estimation in Machine Translation
|
I'm delighted to announce that we have been successful in a number of FP7 projects, including:
- EuroMatrix+: Bringing Machine Translation for European Languages to the User
- Panacea: Platform for Automatic, Normalized Annotation and Cost- Effective Acquisition of Language Resources for Human Language Technologies
- CoSyne: Multi-Lingual Content Synchronisation with WIKIS
- PLuTO: Patent Language Translations Online
- T4ME: Technologies for the Multilingual European Information Society
See our Projects section for details on each individual project.
These projects will bring in another 11 staff to our already large, well-established team. Some of our recent recruits include:
- Dr. Raphael Rubino
- Dr. Xiaofeng Wu
- Dr. Junhui Li
- Rasoul Samad Zadeh Kaljahi
|
| See our People section for further information. Once all the new staff are in place, our MT team here will constitute 33 people. Other project proposals are currently under review. |
| We'll continue to get involved in large-scale MT evaluation campaigns (e.g. we've recently participated in IWSLT, NTCIR and MEDAR campaigns, with WMT and MetricsMATR evaluation teams just been put together).
| | Note also that our determination to make some of our software available in the public domain has resulted in the release of Marclator, a standalone EBMT system which we use as a wrapper around Moses in our successful MT-evaluation campaigns. We hope you try it out, and get back to us with comments! This is intended to be the first release of MT software from our group, with the ultimate intention to release the full MaTrEx system.
We have formed links with some of the leading groups in the fields of MT , where we carry out joint research, host research visits and submit joint research proposals. We would be delighted to become actively involved with more research groups, so contact us if interested.
Members of the team play a prominent role in the European Association for MT and the International Association for MT. We urge you to join your local association! Furthermore, members of the group are leading representatives on the Editorial Board of the journal Machine Translation. Again, we would encourage you to submit papers to the journal, or review books that are of interest to the readership of the journal. |
| Thanks for accessing our site, and let us know if it did or did not meet your expectations -- if there's something wrong, or missing, we can only fix it if you let us know! |
| -- Andy Way, February 2010 |
|
|
| Recent News |
 April 27, 2012 DCLRS Seminar: 3 Guest speakers, Grzegorz A. Kleparski, Marta Pikor-Niedziałek and Edyta Więcławska from University of Rzeszow, Poland at 4pm, Friday, 27th April, in L2.21 School of Computing, DCU. |
| April 25, 2012 - Irish Times publishes article on Sara Morrissey's work on Sign Language Machine Translation system |
| April 13, 2012 - DCLRS Seminar by Ana Guerberof, visiting researcher in SALIS, DCU on Productivity and quality in the post-editing of outputs from translation memories and machine translation at 4pm, Friday, 13th April, in L2.21 School of Computing, DCU. |
| Apr 5, 2012 - 5 papers from our group accepted for publication at the 16th Annual Conference of European Association of Machine Translation(EAMT-2012). Trento, Italy |
|
