Statistical Language Processing and Learning Lab.

Professor Dr Khalil Sima'an
Chair of Computational Linguistics
Language and Computation
Institute for Logic, Language and Computation (ILLC)
Faculty of Science, Universiteit van Amsterdam

---

Our research concentrates on statistical models for structured language processing with application to machine translation, paraphrasing, semantic and morpho-syntactic parsing, and statistical learning for NLP . Machine Translation is currently a central embedding application for our work. We aim at a range of phenomena in the quest for more adequate and more fluent MT systems learned from bilingual parallel data, including reordering, morphological variation, domain adaptation, evaluation and tuning.  The general approach we take aims at inducing the latent structure that represents relevant salient regularities in natural language data (mono- and multi-lingual corpora) for improved language applications. Our current work concentrates on exploiting regularities in word aligned parallel data for learning hierarchical reordering models over permutations and word alignments, e.g., learning hierarchical preordering from bare word alignments. inducing models sensitive to domain variation in big parallel data, e.g., data selection, word alignment, model adaptation. devising better MT evaluation metrics, e.g., BEER. inducing novel semantic representations within meaning-preserving language processing models as a surrogate for actual semantic representations that work with a form and its referent. inducing morpho-syntactic generation models for translation into morphologically-rich languages.

Overview

Researchers

Projects

Alumni

---

Ressearchers@SLPL lab. . Khalil Sima'an (Vici NWO, ILLC, UvA) Principal Investigator . Gideon Wenniger (NWO, Open Comp). Jun 2010 - Jun 2014 PhD student Alignment and Hierarchical SMT . Sophie Arnoult (NWO, Open Comp). Aug 2012 - Aug 2017 PhD student TAG and Hierarchical SMT . Milos' Stanojevic' (STW  DatAptor) Mar 2013 - Mar 2017 PhD student MT and Hierarchical Alignments . Hoang Cuong (EXPERT ITN) Oct 2013 - Oct 2017 PhD student Hierarchical MT with TMs  . Joachim Daiber (EXPERT ITN) Oct 2013 - Oct 2017 PhD student Hierarchical MT with TMs . Philip Schulz (Vici) Nov 2013 - Nov 2017 PhD student MT and Meaning Preserving Models . Aaron Li-Feng Han Sep 2014 - Sep 2017 Research Internship MT Evaluation . Amir Kamran (STW DatAptor) Jan 2014 - Jun 2016 Researcher MT and Domain Adaptation . Wilker Aziz (Vici) Jan 2015 - 1 Jan 2018 Postdoc SMT . Bushra Jawaid (STW DatAptor) Jan 2015 - 1 Jan 2016 Researcher SMT adaptation . Joost Bastings (Vici) Jan 2015 - Dec 2018 PhD student SMT and meaning preservation . Christos Louisos Nov 2014 - Oct 2015 Programmer SMT and adaptation . Stella Frank (EC QT21) Jan 2015 - Dec 2017 postdoc SMT, learning, morpho-syntax and semantics . Vacancy (Vici) Apply postdoc SMT and meaning preservation . Bart Mellebeek (STW  DatAptor) Jan 2013 - Oct 2014 Postdoc MT and Domain Adaptation Ongoing Research projects Grant 2014 QT21, H2020 Cracking the language barrier (3year x 1fte Researcher; Co-applicant and PI UvA, Coordinator DFKI, Germany).  Grant 2013 VICI NWO. Machine Translators: Teaching Computers to Translate Using their own Words (Euro 1.5 million; PI). Grant 2012  Marie Curie ITN project EXPERT (2 PhD positions; Co-applicant and PI UvA, Coordinator University of Wolverhampton/Sheffield) Grant 2012 STW (Technology Foundation) project DatAptor (Euro 750k; PI) Grant 2012 Free Competition of NWO Exact Sciences Board, Statistical Translation of Novel Constructions (Euro 230k; PI) Concluded projects 2010-14 Free Competition of NWO Exact Sciences Board, project proposal: Machine Translation When Exact Pattern Match Fails (Euro 220k) 2006-11 Vidi of NWO Exact Sciences Board: Priors for the Estimation of Stochastic Grammars from Incomplete Natural Language Data (Euro 760k) 2004-07  Free Competition of NWO Exact Sciences Board, project proposal: Learning Stochastic Tree Grammars from Treebanks (Euro 220k) 2002-03  KNAW Fellow (Royal Netherlands Academy for Arts and Sciences) 2000-02  Postdoctoral grant at UvA and Tilburg University, Probabilistic memory-based models of parsing (K. Sima'an, R. Scha and W. Daelemans). Software and Data Packages Beer: MT Evaluation metric available for download (no additives, pure ingredients and straight from the tap! - contact Milos' Stanojevic') Word and Hierarchical Alignment Visualizer Package (contact Gideon Wenniger) FREVAL An All Fragments Parser Evaluation Metric: https://github.com/bastings/freval The Hebrew Treebank Project (with Yoad Winter, Alon Itai and few more). AmbLex Bi-directional Estimation for Noisy-Channel Models (with Markos Mylonakis) dopdis Statistical parsing (covers STSGs and PCFGs -- with contribution from Andreas Zollmann). MorphTagger A Segmenter + POS tagger for Modern Hebrew (with Roy Bar-Haim, Yoad Winter and Alon Itai).

---

Former Ph.D. students Joachim Daiber: March 2018 Milos Stanojevic: 13 December 2017 Hoang Cuong: graduation 6 July 2017 Gideon Maillette de Buy Wenniger: graduation 10 June 2016 Markos Mylonakis (UvA, NWO VIDI): graduation 19 January 2012 Reut Tsarfaty (UvA, NWO MOSAIEK project): 2011 Hany Hassan co-supervision together with  Andy Way  at Dublin City University, Dublin, Ireland.  2011 Former postdocs Detlef Prescher (2003-2006 -- Project LeStoGram Consistent Statistical Estimation from Treebanks) Yoav Seginer -- substituting  lecturer for my statistical NLP courses (2007--2008) Tejaswini Deoskar -- 50% substituting lecturer and 50% postdoc (2008-10, VIDI project, NWO) Maxim Khalilov (2009-2011) Postdoc on the VIDI project Bart Mellebeek (STW  DatAptor),  postdoc, Jan 2013 - Oct 2014. Desmond Elliot (2016 - 2017) Stella Frank (2016 - 2017) Former M.Sc. students Name Where and When Subject After graduation went to... Jakub Zavrel U. Utrecht, 1995/6 Vector-Space Models for Parsing TextKernel (co-founder) Jorn Veenstra U. Utrecht, 1995/6 - with Joos Kok Head Correlation Detection for Syntactic Analysis Consultant Vera Hollink UvA 2002 - with Henk Zeevat   Anaphora Resolution by Probabilistic Parsers UvA, Amsterdam (PhD student, graduated) Luciano Buratto UvA 2002    DOP Estimation by Backoff Smoothing            MoL-2002-07 U. of Warwick, UK (PhD student, graduated) Oren Tsur UvA 2003 - with Maarten de Rijke QA and Learning Bibliography Classifiers       MoL-2003-06 Hebrew University, Israel (PhD student, graduated) Andreas Zollmann UvA 2004 - with Detlef Prescher A Consistent and Efficient Estimator for DOP  MoL-2004-02 CMU, USA (PhD student, graduated) Roy Bar-Haim Technion 2004/5 - with Y.Winter+ A. Itai Probabilistic Methods for Hebrew Morphological Analysis     U. of Bar-Ilan, Israel (PhD student, graduated) Thuy Linh Nguyen UvA 2004 Rank-Consistency and DOP Estimation CMU, USA (PhD student, graduated) Felix Hageloh UvA 2006 Simulating Collins'97 model using Treebank Transforms and PCFGs Softwar Engineer Markos Mylonakis UvA 2006/7 Bi-directional Noisy-Channel Estimators (Bi-EM) UvA, ILLC (PhD student, graduated) Barbara Plank UvA 2007 Parsing with Domain-awareness Groningen University (Ph.D. Student, graduated) Saib Mansour Technion 2008 withA. Itai and Yoad Winter Segmentation and POS tagging for Arabic and Hebrew RWTH Aachen (Ph.D. student, graduated) Sanne Korzec UvA 2010 Phrase probability estimation in SMT Industry Sophie Arnoult UvA 2011 Adjuncts in Statistical Machine Translation Univ. of Amsterdam, PhD student Joost Bastings UvA 2012 Parsing with graph symbols SAAB Sweden Consultant Katya Garmash UvA 2012 Paraphrasing and SMT (ongoing) University of Amsterdam, PhD student Dieuwke Hupkes UvA 2013 Translation equivalence and Syntactic Structure University of Amsterdam, Research Assistant