Andy Way's Home Page


Publications

Teaching

Research

Past Postgraduate Students

School Staff Phonebook

Webmail

DCU MT Group

MT Archive


Andy Way, B.A., M.Sc., Ph.D.


Associate Professor in Computing
Address: School of Computing, Dublin City University, Glasnevin, Dublin 9, IRELAND.

Tel: +353-1-7005644, Fax: +353-1-7005442, Email: away@computing.dcu.ie

My little Chelsea supporters!


I am an Associate Professor in the School of Computing at Dublin City University, and have been here since 1991. I am also the School Research Convenor.

I am the Editor for the journal Machine Translation. Please contact me if you would like to submit to the journal.

I am also on the EAMT Committee, from the EAMT-09 conference as the new President of EAMT. DCU also act as Webmaster for EAMT.

I am also a member of the National Centre for Language Technology (NCLT), and am a member of the Language and Intelligence Research group in our School.

I am also the track leader for Integrated Language Technologies in the Centre for Next Generation Localisation, aiming at facilitating optimal multilingual applications for deployment in the localisation industry.


Recent News

We've just heard that two of our papers have been accepted for publication in Machine Translation. The first paper is entitled Bidirectional Data-Driven Machine Translation for Irish and German Sign Languages, and represents joint work with Sara Morrissey. The second paper is entitled Parallel Treebanks and their Exploitability in Machine Translation, and represents joint work with John Tinsley.

We've just had a paper accepted for presentation at Interspeech 2009, the 10th Annual Conference of the International Speech Communication Association, which will take place on 6-10 September, 2009 in Brighton, UK. The paper is entitled Using Same-Language Machine Translation to Create Alternative Target Sequences for Text-To-Speech Synthesis, and is joint work with Peter Cahill and Julie Berndsen of UCD, and Jinhua Du of DCU.

We've just had four papers accepted for presentation at MT Summit 2009, the Twelfth Machine Translation Summit, organized by the International Association for Machine Translation and the Association for Machine Translation in the Americas, which will be held at the Château Laurier, Ottawa, Canada, 26-30 August 2009. The papers are as follows:

  • Source-Side Context-Informed Hypothesis Alignment for Combining Outputs from Machine Translation Systems: Jinhua Du and Andy Way
  • Tracking Relevant Alignment Characteristics for Machine Translation: Patrik Lambert, Yanjun Ma, Sylwia Ozdowska and Andy Way
  • Improving the Objective Function in Minimum Error Rate Training: Yifan He and Andy Way
  • Using Percolated Dependencies for Phrase Extraction in SMT: Ankit Srivastava and Andy Way

We've just had a paper accepted for presentation at RANLP 2009, the International Conference on Recent Advances in Natural Language Processing, to take place September 14-16, 2009, in Borovets, Bulgaria. The paper is entitled Lexicalized Semi-Incremental Dependency Parsing and is joint work with Hany Hassan and Khalil Sima'an.

We've just had two papers accepted for presentation at EMNLP, which will be held on August 6--7 at the Suntec Singapore International Convention & Exhibition Centre, immediately following ACL/IJCNLP 2009. The first paper is entitled A Syntactified Direct Translation Model with Linear-time Decoding and is joint work with Hany Hassan and Khalil Sima'an. The second paper is entitled Accuracy-Based Scoring for DOT: A Step Towards Evaluation Measure Based MT Training and is joint work with Sergio Penkale and Daniel Galron, a student of Dan Melamed's who recently completed a placement in DCU funded by CNGL.

We've just had a paper accepted for presentation at the Named Entities Workshop (NEWS) 2009, to take place on 7th August, 2009 at ACL/IJCNLP 2009. The paper is entitled English--Hindi Transliteration Using Context-Informed PB-SMT: the DCU System, and is joint work with Rejwanul Haque, Ankit Srivastava and Sudip Naskar.

I'm on the programme committee for IWSLT 2009, which will be held in Tokyo, Japan, on December 1-2, 2009. This year IWSLT-09 will be co-located with IUCS 2009.

At the EAMT-09 conference in Barcelona, I was elected as the new President of EAMT, following the sterling service provided to our community in this regard by Bente Maegaard.

We've just had a paper accepted for presentation at SETQA-NLP 2009, a workshop at HLT-NAACL 2009, on "Software engineering, testing, and quality assurance for natural language processing". This will take place in Boulder, Colorado, on Friday, June 5th. The paper is entitled Web service integration for next generation localisation, and represents joint work with David Lewis, Stephen Curran, Kevin Feeney, Zohar Etzioni, John Keeney (all TCD) and Reinhard Schäler (University of Limerick).

I'm on the programme committee for EMNLP, which will be held on August 6--7 at the Suntec Singapore International Convention & Exhibition Centre, immediately following ACL/IJCNLP 2009.

I'm on the programme committee for MT Summit XII, which will take place from August 25--29 2009 at the Chateau Laurier in Ottawa, Ontario, Canada.

We've just had five papers accepted for presentation at EAMT-09, the 13th Annual Meeting of the European Association for Machine Translation, which will take place May 14-15 at UPC in Barcelona, Spain. The five papers are as follows:

  • Using Supertags as Source Language Context in SMT: Rejwanul Haque, Yanjun Ma, Sudip Naskar, Andy Way
  • Tuning Syntactically Enhanced Word Alignment for Statistical Machine Translation: Yanjun Ma, Patrik Lambert, Andy Way
  • Learning Labelled Dependencies in Machine Translation Evaluation: Yifan He and Andy Way
  • Marker-based Filtering of Bilingual Phrase Pairs for SMT: Felipe Sanchez-Martinez and Andy Way
  • Optimal Bilingual Data for French-English PB-SMT: Sylwia Ozdowska and Andy Way

I'm giving a keynote at the 2nd International Conference on Arabic Language Resources and Tools run by the MEDAR project, to take place on 22-23 April 2009 at the Hotel Grand Hyatt Cairo, Egypt.

We've just had a paper accepted for presentation in the Special Issue on Machine Translation of Asian Languages of ACM TALIP. The paper is entitled Bilingually Motivated Word Segmentation for Statistical Machine Translation, and is joint work with Yanjun Ma.

We've just had a paper accepted for presentation at the Fourth Workshop on Statistical Machine Translation, EACL 2009, to take place in Athens, Greece, on March 30 and 31, 2009. The paper is entitled MaTrEx: the DCU MT System for WMT 2009, and is joint work with Jinhua Du, Yifan He and Sergio Penkale.

We've just had a paper accepted for presentation at CICLING 2009, to be held in Mexico City from March 1-7, 2009. The paper is entitled Parallel Treebanks in Phrase-Based Statistical Machine Translation and is joint work with John Tinsley and Mary Hearne.

We've just had a paper accepted for presentation at EACL-09 to be held in Athens, Greece, from March 30-April 3, 2009. The paper is entitled Bilingually Motivated Domain-Adapted Word Segmentation for Statistical Machine Translation and is joint work with Yanjun Ma.

I'm on the Programme Committee for the Joint ACL-IJCNLP 2009 conference to be held in Singapore on August 2-7, 2009.

We've just submitted our system description paper for the NLP Tools Contest: Statistical Machine Translation (English to Hindi), part of the 6th International Conference on Natural Language Processing in Pune, India, to be held on 20-22 December 2008.

I'm attending the JHU 2009 Summer Workshop Planning Meeting this weekend Nov 7--9. Khalil Sima'an and I have put together an proposal to incorporate syntax into today's PB-SMT systems. We'll see how we get on ...

I'm on the Programme Committee for the 12th Conference of the European Chapter of the Association for Computational Linguistics (EACL-09) to be held in Athens, Greece, from March 30-April 3, 2009.

New recruits to our team include Marianna Apidianiki (Jussieu, Paris: postdoc), Anton Bryl (Trento: postdoc), Yifan He (Tsinghua Univ., Beijing: PhD), Rejwanul Haque (Jadavpur Univ., Kolkata: PhD), and Sandipan Dandapat (IIT-Hyderabad: PhD). Welcome all!

We've just had a paper accepted for presentation at the Second IEEE Spoken Language Technology (SLT) workshop to be held in Goa, India in December 2008. The paper is entitled A Syntactic Language Model based on Incremental CCG Parsing, and is joint work with Hany Hassan and Khalil Sima'an.

I'm on the organizing committee (as one of the publicity chairs) for ACL-IJCNLP to take place in Singapore 2-7 August 2009.

I'm on the Programme Committee for the 8th Conference of the Association for Machine Translation in the Americas (AMTA-08) to be held in Waikiki, Hawaii, from October 21-25, 2008.

I'm delighted to state that Dr. Jinhua Du has joined us from the Chinese Academy of Science in Beijing as a post-doctoral researcher. Jinhua started on July 1st, and will be with us for a year on the Next Generation Localisation CSET project.

I'm pleased to announce that Sudip Naskar from the Computer Science and Engineering Dept. at Jadavpur University in Kolkata, India, will be joining us on the Prospect MT project as a postdoctoral researcher from July 1st.

We've just had a paper accepted for presentation at Coling 2008 to take place in Manchester from August 18-22. The paper is entitled Automatic Generation of Parallel Treebanks, and is joint work with Ventsislav Zhechev.

I'm delighted to announce that Mikel Forcada of the Universitat d'Alacant in Spain has successfully applied for an SFI Walton scholarship, to enable him to spend a year with our MT group from June 2009.

Sara Morrissey has successfully defended her PhD (2nd May). See here for more details.

I'm on the Programme Committee for the Second International Symposium on Universal Communication to be held in Osaka, Japan from December 15-16 2008.

Karolina Owczarzak has successfully defended her PhD (25th April). See here for more details.

I'm on the Programme Committee for the Second IEEE Spoken Language Technology (SLT) workshop to be held in Goa, India in December 2008.

We had a paper accepted for presentation at the 2nd SSST workshop at ACL-08, to take place in June in Columbus, OH. The paper is entitled Improving Word Alignment Using Syntactic Dependencies, and is joint work with Yanjun Ma, Yanli Sun, and Sylwia Ozdowska.

I'm delighted to state that Dr. Patrik Lambert has joined us from the UPC in Barcelona as a post-doctoral researcher. Patrik started on April 1st, and will be with us for a year on the Next Generation Localisation CSET project.

An article of ours has just appeared in Computational Linguistics. It's entitled Wide-Coverage Deep Statistical Parsing Using Automatic Dependency Structure Annotation, and is joint work with Aoife Cahill, Mick Burke, Ruth O'Donovan, Stefan Riezler, and Josef Van Genabith.

We've just had a paper accepted for publication in Machine Translation. The paper is entitled Evaluating Machine Translation with LFG Dependencies, and is joint work with Karolina Owczarzak and Josef Van Genabith.

We've just had a paper accepted for presentation at LREC-08. The paper is entitled The ATIS Sign Language Corpus, and represents joint work with Jan Bungeroth, Daniel Stein, Hermann Ney, Sara Morrissey, and Lynette van Zijl.

I'm on the Programme Committee for IWSLT 2008, to take place in (ahem!) Hawaii on October 20--21. IWSLT is co-located with AMTA 2008 this year.

I'm a keynote speaker at the 2nd Symposium on Innovations in Machine Translation Technologies taking place in Tokyo between March 19--21. This is hosted by the National Institute of Information and Communications Technology (NICT).

I'm on the programme committee for the SSST workshop at ACL-08, to take place in June in Columbus, OH.

I reviewed the EuroMatrix project in Prague Feb 04-05.

I gave a talk at the Mixing Approaches to Machine Translation workshop in Donostia in the Basque Country on Feb 14.

Items from 2007

We had a paper accepted for Sixth International Workshop on Treebanks and Linguistic Theories (TLT-07), to take place in Bergen, Norway, in December 07. The paper is entitled Exploiting Parallel Treebanks to Improve Phrase-Based Statistical Machine Translation, and is joint work with John Tinsley and Mary Hearne.

I presented our system's results at IWSLT-07 in Trento, Italy, on Oct 15th.

I'm delighted to state that Sylwia Ozdowska has joined us from the University of Toulouse as a post-doctoral researcher. Sylwia started on Sept 1st, and will be with us for two years on the PROSPECT project.

Bart Mellebeek recently successfully defended his PhD. See here for more details.

Itzulpen automatikoaren erronka handiena kalitatea da!

The list of accepted papers for TMI 07 is now available.

We had five papers accepted for TMI 07 to be held in Skövde, Sweden, from September 7-9. The papers are entitled Exploiting Source Similarity for SMT using Context-Informed Features (with Nicolas Stroppa and Antal van den Bosch), Capturing Translational Divergences with a Statistical Tree-to-Tree Aligner (with Mary Hearne, John Tinsley and Ventsislav Zhechev), Hand in Hand: Automatic Sign Language to Speech Translation (with Daniel Stein, Philippe Dreuw, Hermann Ney, and Sara Morrissey), A Cluster-Based Representation for Multi-System MT Evaluation (by Nicolas Stroppa and Karolina Owczarzak), and Alignment-Guided Chunking (with Yanjun Ma and Nicolas Stroppa).

We had a paper accepted for the Conference and Workshop on Assistive Technologies for People with Vision and Hearing Impairments - Assistive Technology for All Ages, to take place in Granada, Spain, in August 2007. The paper is entitled Joining Hands: Developing a sign language machine translation system with and for the Deaf Community, and is joint work with Sara Morrissey.

We had three papers accepted for the MT Summit XI in Copenhagen in September 2007. The papers are entitled Comparing Rule-Based and Data-Driven Approaches to Spanish-to-Basque Machine Translation (with Gorka Labaka and Kepa Sarasola from the University of the Basque Country, and Nicolas Stroppa), Robust Language Pair-Independent Sub-Tree Alignment (with Mary Hearne, John Tinsley and Ventsislav Zhechev), and Towards a Hybrid Data-Driven MT System for Sign Language Translation (with Sara Morrissey, and Daniel Stein, Jan Bungeroth, and Hermann Ney from RWTH Aachen).

We have a paper accepted for the 2nd Workshop on SMT at ACL 2007, to take place in Prague, in June. The paper is entitled Labelled Dependencies in Machine Translation Evaluation, and is joint work with Karolina Owczarzak and Josef Van Genabith.

We've just announced the 2nd call for papers for TMI 07 to be held in Skövde, Sweden, from September 7-9, just before MT Summit XI in Copenhagen.

I'm on the programme committee for MT Summit 2007, to be held in Copenhagen from September 10-14, right after TMI in in Skövde, Sweden (September 7-9).

I'm on the programme committee for IWSLT 2007, to be held in Trento on 15-16 October.

We have two papers accepted for ACL 2007, to take place in Prague, in June. The first is entitled Integrating Supertags into Phrase-based Statistical Machine Translation (with Hany Hassan and Khalil Sima'an) and the second is called Bootstrapping Word Alignment Via Word Packing with Yanjun Ma and Nicolas Stroppa.

I'm on the programme committee for the MT track in EMNLP, which directly follows ACL 2007.

We have a paper accepted for the Syntax and Structure in Statistical Translation (SSST) workshop at NAACL-HLT 2007. It is entitled Dependency-Based Automatic Evaluation for Machine Translation (with Karolina Owczarzak and Josef Van Genabith).

I was programme chair for the track on MT at ACL-07, to take place in Prague June 24--29th, 2007.

Much more important than all this MT-related nonsense, our 3rd child Rachel Amy arrived safely at 1.04 am Weds 8th November, 7lb 12 oz (3.5kg). All delighted!

As programme chair, I announced the 1st call for papers for TMI 2007, to be held in Skövde, Sweden, from September 7-9, just before MT Summit XI in Copenhagen.

We have a paper accepted for the METIS-II Workshop on New Approaches to Machine Translation to be held in Leuven on January 11th, 2007. The paper is entitled A memory-based classification approach to marker-based EBMT, and is joint work with Antal van den Bosch and Nicolas Stroppa.

I was on the programme committee for HLT-NAACL 2007.

Older Items

Old News from 2006, Older News from 2004 & 2005, and Even Older News from 2003 (hardly 'news' at all now!)!



Andy Way, 18th June, 2009.