Andy Way's Home Page


Useful Links

Publications

Teaching

Research

Past Postgraduate Students

School Staff Phonebook

DCU MT Group

MT Archive


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

Contact Details

Full Professor in Computing
Address: ADAPT Centre, School of Computing, Dublin City University, Glasnevin, Dublin 9, IRELAND.

Tel: +353-1-7005074, Fax: +353-1-7006702, Email: away@computing.dcu.ie


Between 2011--13, I spent a period of sabbatical leave working in the translation and localisation industry in the UK. From the start of 2014, I have been back in DCU full-time as Professor in the School of Computing at Dublin City University, where I began working in 1991. From January 1st 2014, I became Deputy Director of the CNGL Centre for Intelligent Content at DCU. This programme ceased at the end of August 2015, to be replaced by the ADAPT Centre for Digital Content Technology, where I remain Deputy Director. We received €24 million for 6 years' direct funding of the new ADAPT Centre. Together with industry investment as well as indirect funding from other sources, over €110 million will be invested in this ambitious new research programme.

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 President. My six-year stewardship ended at the 2015 conference in Turkey, with Mikel Forcada taking over.

Following MT Summit 2011, I became President of the International Association for Machine Translation, ending my term of office on 6th September 2013 at Machine Translation Summit XIV in Nice.


Recent News

We've had a paper accepted for presentation at AMTA 2016, to take place in Austin, Texas from Friday, October 28 through Tuesday, November 1, 2016. The paper is entitled Improving KantanMT Training Efficiency with FastAlign and represents joint work with our industry partner KantanMT in the ADAPT Centre.

We'll be presenting a paper at CoNLL 2016 in Berlin this week. The paper is entitled Identifying Temporality of Word Senses Based on Minimum Cuts, and representes joint work with Mohammed Hasanuzzaman, Gaël Dias, Stéphane Ferrari and Yann Mathet.

We've had a paper accepted for the Hermes: Journal of Language and Communication in Business. The paper is entitled On the Complementarity between Human Translators and Machine Translation and is co-authored with Maite Aragonés Lumeras from WIPO.

We've had a paper accepted for the Prague Bulletin of Mathematical Linguistics. The paper is entitled FaDA: Fast Document Aligner using Word Embedding and represents joint work with Pintu Lohar, Debasis Ganguly, Haithem Afli and Gareth Jones.

We'll be presenting a Pilot on Machine-assisted Translation of Novels with Post-editing at the 8th EST Congress in Aarhus, Denmark, taking place from 15th to 17th September, 2016. This is joint work with Antonio Toral.

We've just presented two papers at ACL-2016: The 54th Annual Meeting of the Association for Computational Linguistics, taking place in Berlin, Germany, on August 7--12, 2016. The papers are entitled:

  • Phrase-Level Combination of SMT and TM Using Constrained Word Lattice, and
  • Graph-Based Translation Via Graph Segmentation.
Both papers represent joint work with Liangyou Li and Qun Liu.

I'm on a number of committees for 2016 (and already, 2017!) conferences and workshops, including:

We presented a paper at the Fifth Workshop on Hybrid Approaches to Translation (HyTra) which took place in Riga, Latvia on June 1, 2016. The paper is entitled Improving Subject-Verb Agreement in SMT and represents joint work with Eva Vanmassenhove and Jinhua Du.

Note: I also gave a keynote at this event on Hybrid MT in Real-World Applications.

We had a paper accepted for publication in Transactions on Asian and Low-Resource Language Information Processing. The paper is entitled Boosting Neural POS Tagger for Farsi Using Morphological Information and represents joint work with Peyman Passban and Qun Liu.

The ADAPT Centre MT team were very successful in having had seven papers accepted for presentation at EAMT 2016, which took place in Riga, Latvia from May 30 - June 1, 2016. The papers are as follows:

  • Research Track:
    • Joss Moorkens and Andy Way. Comparing Translator Acceptability of TM and SMT Outputs
    • Liangyou Li, Carla Parra Escartín and Qun Liu. Combining Translation Memories and Syntax-Based SMT: Experiments with real industrial data
    • Peyman Passban, Chris Hokamp, Andy Way and Qun Liu. Improving Phrase-Based SMT Using Cross-Granularity Embedding Similarity
  • User Track:
    • Ke Hu. A Comparative Study of Post-editing Guidelines
    • Antonio Toral, Raphael Rubino and Gema Ramírez-Sánchez. Re-assessing the Impact of SMT Techniques with Human Evaluation: a Case Study on English--Croatian
  • Product Track:
    • Valia Kordoni, Joss Moorkens and Andy Way. TraMOOC (Translation for Massive Open Online Courses): Providing Reliable MT for MOOCs
    • Antonio Toral, Sergio Ortiz Rojas, Mikel Forcada, Nikola Ljubešić and Prokopis Prokopidis. Abu­MaTran: Automatic building of Machine Translation

Our paper entitled From Arabic User-Generated Content to Machine Translation: Integrating Automatic Error Correction by Haithem Afli, Walid Aransa, Pintu Lohar and myself won the best presentation award at CICLING 2016: 17th International Conference on Intelligent Text Processing and Computational Linguistics, which took place 3rd to 9th April in Konya, Turkey.

A number of special issues of the Machine Translation journal have been announced, namely:

  • Neural Network Approaches to Machine Translation, co-edited by Xiaodong He (Microsoft Research, US), Qun Liu (ADAPT Centre, Dublin City University, Ireland) and Hermann Ney (RWTH Aachen, Germany). Deadline: April 30th 2016.
  • Natural Language Processing for Translation Memories, co-edited by Constantin Orasan (University of Wolverhampton, UK) and Marcello Federico (FBK, Italy). Deadline: May 15th 2016.
  • Spoken Language Translation, co-edited by Alex Waibel (Carnegie Mellon University / Karlsruhe Institute of Technology), Sebastian Stüker (Karlsruhe Institute of Technology), Marcello Federico (Fondazione Bruno Kessler), Satoshi Nakamura (Nara Institute of Science and Technology), Hermann Ney (RWTH Aachen University) and Dekai Wu (The Hong Kong University of Science and Technology). Deadline: July 15th 2016.
If you are interested in submitting to any of these special issues, please contact me or the co-editors responsible for each topic. Of course, we welcome submissions for regular issues of the journal, too! We also have a 'fast-tracking' option for extended versions of papers that have appeared at peer-reviewed conferences to critical acclaim!

Wepresented a paperat the 6th International Workshop on Controlled Language Applications, which took place at LREC 2016 in Portorož, Slovenia. The paper is entitled ProphetMT: Controlled Language Authoring Aid System Description and represents joint work with Xiaofeng Wu, Liangyou and Jinhua Du.

We presented a paper at ETHI-CA² 2016: ETHics In Corpus Collection, Annotation & Application, which took place at LREC 2016 in Portorož, Slovenia. The paper is entitled Translation Resources and Translator Disempowerment and represents joint work with Joss Moorkens, David Lewis, Wessel Reijers, and Eva Vanmassenhove in the ADAPT Centre.

We presented a paper at NAACL HLT 2016: The 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, which took place in San Diego, California, June 12--17, 2016. The paper is entitled A Novel Approach for Dropped Pronoun Translation, and represents joint work with Longyue Wang, Zhaopeng Tu, Xiaojun Zhang, Hang Li,and Qun Liu, extending the excellent collaboration between the Huawei Noah's Ark Lab and the ADAPT Centre.

We presented two papers at CICLING 2016: 17th International Conference on Intelligent Text Processing and Computational Linguistics, which took place from 3rd to 9th April in Konya, Turkey. The papers are as folows:

  • Improved Named Entity Recognition using Machine Translation-based Cross-lingual Information: Sandipan Dandapat and Andy Way
  • From Arabic User-Generated Content to Machine Translation: Integrating Automatic Error Correction: Haithem Afli, Walid Aransa, Pintu Lohar and Andy Way

As National Anchor Point in Ireland for the European Language Resource Coordination (ELRC), we held our national CEF Workshop in Dublin on January 28th. The event was very successful, and thanks to all participants for attending, especially those who demonstrated interest in sharing public data to improve the training of our English<=>Irish MT engines! There's still plenty of time to get involved, so even if you missed out on the event, please do get in touch!

The ADAPT Centre team at DCU enjoyed remarkable success in getting ten papers accepted for the LREC 2016 main conference, which took place in Portorož, Slovenia from 23-28 May 2016. The list of accepted papers is as follows:

  • TweetMT: A parallel microblog corpus: Iñaki San Vicente, Iñaki Alegria, Cristina España-Bonet, Pablo Gamallo, Hugo Gonçalo Oliveira, Eva Martinez Garcia, Antonio Toral and Arkaitz Zubiaga
  • Evaluating the Impact of Light Post-Editing on Usability: Sheila Castilho and Sharon O'Brien
  • ProphetMT: A Tree-based SMT-driven Controlled Language Authoring/Post-Editing Method: Xiaofeng Wu, Jinhua Du, Qun Liu and Andy Way
  • The Automatic Construction of Discourse Corpus for Dialogue Translation: Longyue Wang, Xiaojun Zhang, Zhaopeng Tu, Andy Way and Qun Liu
  • Investigation on OCR Error Correction for Historical Texts: Haithem Afli, Zhengwei Qiu, Andy Way and Páraic Sheridan
  • Towards producing bilingual lexica from monolingual corpora: Jingyi Han and Nuria Bel
  • Using BabelNet to Augment Statistical Machine Translation: Jinhua Du, Andy Way and Andrzej Zydron
  • Enhancing Access to Online Education: Quality Machine Translation of MOOC Content: Valia Kordoni, Antal van den Bosch, Katia Lida Kermanidis, Vilelmini Sosoni, Kostadin Cholakov, Iris Hendrickx, Matthias Huck and Andy Way
  • Developing a Dataset for Evaluating Approaches for Document Expansion with Images: Iacer Calixto, Debasis Ganguly and Gareth Jones
  • Enhancing cross-border EU e-commerce through Machine Translation: needed Language Resources, challenges and opportunities: Meritxell Fernandez, Vladimir Popescu, Antonio Toral and Khalid Choukri
This represents work done on the core ADAPT grant, EU projects (MLi, Falcon and TraMOOC), as well as research carried out by visiting researchers in our MT lab. Well done to all!

We've just presented a paper at the Global Wordnet Conference 2016 in Bucharest, Romania, the conference taking place from 27--30 January 2016. The paper is entitled Using Wordnet to Improve Reordering in Hierarchical Phrase-Based Statistical Machine Translation, and represents joint work with Arefeh Kazemi and Antonio Toral.

I've recently been appointed National Anchor Point in Ireland for the European Language Resource Coordination (ELRC), whose objective is to identify and gather language and translation data relevant to national public services, administrations and governmental institutions across all 30 European countries participating in the Connecting Europe Facility (CEF) Programme.

Less Recent News

We had two papers accepted for MT Summit in Miami, Florida, which took place at the end of October 2015. The papers are as follows:
  • An Empirical Study of Segment Prioritization for Incrementally Retrained Post-Editing-Based SMT: Jinhua Du, Ankit Srivastava, Andy Way, Alfredo Maldonado-Guerra and David Lewis (Main track)
  • Domain Adaptation for Social Localisation-based SMT: A Case Study Using Trommons Platform: Jinhua Du, Andy Way, Zhengwei Qiu, Asanka Wasala and Reinhard Schäler (WPTP 2015)

Our paper has appeared in Translation Spaces. The paper is entitled Machine Translation to assist with the Translation of Literary Text and represents joint work with Antonio Toral.

We had a paper accepted for presentation at EMNLP 2015, which took place in Lisbon, Portugal, from September 17--21 2015. The paper was entitled Dependency Graph-to-String Translation and represents joint work with Liangyou Li and Qun Liu.

We had two papers accepted for presentation at WMT 2015, which took place in Lisbon, Portugal, on 17--18 September 2015. The papers are as follows:

  • ParFDA for Fast Deployment of Accurate Statistical Machine Translation Systems, Benchmarks, and Statistics: Ergun Bicici, Qun Liu and Andy Way
  • Referential Translation Machines for Predicting Translation Quality and Related Statistics: Ergun Bicici, Qun Liu and Andy Way

We had a number of papers accepted for EAMT 2015, which took place in Antalya, Turkey, from May 11--13th. The papers were as follows:

  • Dependency-based Reordering Model for Constituent Pairs in Hierarchical SMT: Arefeh Kazemi, Antonio Toral, Andy Way, Amirhassan Monadjemi and Mohammadali Nematbakhsh
  • Post-Editing Evaluations: Trade-offs between Novice and Professional Participants: Joss Moorkens and Sharon O'Brien
  • Benchmarking SMT Performance for Farsi Using the TEP++ Corpus: Peyman Passban, Andy Way and Qun Liu
  • The potential and limits of lay post-editing in an online community: Linda Mitchell
We also presented five projects during the EAMT conference:
  • Abu-MaTran: Automatic building of Machine Translation: Antonio Toral, Tommi A Pirinen, Andy Way, Gema Ramírez-Sánchez, Sergio Ortiz Rojas, Raphael Rubino, Miquel Esplà, Mikel Forcada, Vassilis Papavassiliou, Prokopis Prokopidis and Nikola Ljubešić
  • Tapadóir: Eimear Maguire, John Judge and Teresa Lynn
  • TraMOOC: Translation for Massive Open Online Courses: Valia Kordoni, Kostadin Cholakov, Markus Egg, Andy Way, Lexi Birch, Katia Kermanidis, Vilelmini Sosoni, Dimitrios Tsoumakos, Antal van den Bosch, Michael Papadopoulos, Panayota Georgakopoulou, Maria Gialama, Menno van Zaanen, Gerald Hiller, Ioana Buliga, Mitja Jermol and Davor Orlic
  • FALCON: Federated Active Linguistic data CuratiON: David Lewis
  • HandyCAT - An Open-Source Platform for CAT Tool Research: Christopher Hokamp and Qun Liu

We had a paper accepted for the Fourth Workshop on Computational Linguistics for Literature, colocated with NAACL, to take place on June 4th, 2015 in Denver, Colorado, USA. The paper is entitled Translating Literary Text between Related Languages using SMT, and represents joint work with Antonio Toral.

I was also on a number of committees for 2015 conferences, including:

Andy Way, 17th August, 2016.