Graph and Tree Transducer Interpreter and Development Environment
Publications
PhD Thesis
B. Bohnet (2006) Textgenerierung durch Transduktion linguistischer Strukturen. DISKI 298. AKA, Berlin.
2012
Bernd Bohned and Joakim Nivre. A Transition-Based System for Joint Part-of-Speech Tagging and Labeled Non-Projective Dependency Parsing. EMNLP 2012.
Bernd Bohnet, Anders Björkelund, Jonas Kuhn, Wolfgang Seeker and Sina Zarriess. Generating Non-Projective Word Order in Statistical Linearization. EMNLP 2012.
Richrd Farkas and Bernd Bohnet. Stacking of Dependency and Phrase Structure Parsers. COLING 2012
Wolfgang Seeker, Richard Farkas, Bernd Bohnet, Helmut Schmid and Jonas Kuhn. Data-driven Dependency Parsing With Empty Heads. COLING 2012.
Leo Wanner, Simon Mille, Bernd Bohnet. 2012. Towards a Surface Realization-Oriented Corpus Annotation. INLG 2012.
Bernd Bohnet and Jonas Kuhn. 2012. The Best of Both Worlds - A Graph-based Completion Model for Transition-based Parsers. EACL. 2012, 77-87
Anja Belz, Bernd Bohnet, Simon Mille, Leo Wanner and Michael White. 2012. The Surface Realisation Task: Recent Developments and Future Plans. In Proceedings of the 7th International Natural Language Generation Conference. INLG 2012.
2011
Bernd Bohnet. 2011. Advanced Graph-based and Transition-based Dependency Parsers. International Conference on Dependency Linguistic, 282-290
Bernd Bohnet, Leo Wanner, and Simon Mille. 2011. Statistical Language Generation from Semantic Structures. International Conference on Dependency Linguistic, 251-262
Richard Farkas, Bernd Bohnet, Helmut Schmid: Features for
Phrase-Structure Reranking from Dependency Parses. IWPT 2011: 209-214
Bernd Bohnet, Simon Mille, Benoite Favre, Leo Wanner 2011. StuMaBa: From Deep representation to Surface. ENLG 2011. Surface-Generation Shared Task. Nancy, France.
2010
Bernd Bohnet, Leo Wanner, Simon Mill and Alicia Burga. 2010. Broad Coverage
Multilingual Deep Sentence Generation with a Stochastic Multi-Level
Realizer. To appear at The 23rd International Conference on Computational Linguistics
(COLING 2010), Beijing, China.
Wolfgang Seeker, Bernd Bohnet, Lilja Ovrelid and Jonas Kuhn. 2010. Informed ways of
improving data-driven dependency parsing for German. The 23rd International Conference on Computational Linguistics
(COLING 2010), Beijing, China.
Bernd Bohnet. 2010. Top Accuracy and Fast Dependency Parsing is not a Contradiction. The 23rd International Conference on Computational Linguistics
(COLING 2010), Beijing, China.
Bernd Bohnet and Leo Wanner. (2010) Open Soucre Graph Transducer
Interpreter and Grammar Development Environment. Proceedings of the Seventh
conference on International Language Resources and Evaluation (LREC'10).
Leo Wanner, Bernd Bohnet, Nadjet Bouayad-Agha, Francois Lareau, Daniel Nicklaß. 2010.
Marquis: Generation of User-Tailored Multilingual Air Quality Bulletins. Applied Artificial Intelligence. 24(10)
Benoit Favre, Bernd Bohnet, and Dilek Hakkani-Tür. (2010) Evaluation of Semantic Role
Labeling and Dependency Parsing of Automatic Speech Recognition Output. 35th
International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
2009
Benoit Lavre and Bernd Bohnet. (2009) The Generation of References to the Main Subject and
Named Entities using Conditional Random Fields. ACL-IJCNLP 2009 Workshop: Language Generation and Summarisation (UCNLG+Sum).
Bernd Bohnet. (2009) Synchronous Parsing of Syntactic and Semantic
Structures. Proceedings of 4th International Conference on Meaning-Text Theory.
Bernd Bohnet. (2009) Efficient Parsing of Syntactic and Semantic Dependency
Structures. Proceedings of Conference on Natural Language Learning (CoNLL),
Boulder, 67-72.
2008
Bernd Bohnet. (2008) IS-G. Comparison of Different Learning Techniques for the
Selection of the Main Subject Reference. Proceedings of the Fifth International
Natural Language Generation Conference, Columbus, 192-194
Bernd Bohnet. (2008) The Fingerprint of Human Referring Expressions and their Surface
Realization with Graph Transducers. Proceedings of the Fifth International
Natural Language Generation Conference, Columbus, Ohio, 207-211
Gabriele Taentzer, Enrico Biermann, Denes Bisztray, Bernd Bohnet, Iovka
Boneva, Artur Boronat, Leif Geiger, Rubino Geiss, Akos Horvath, Ole Kniemeyer,
Tom Mens, Benjamin Ness, Detlef Plump, Tamas Vajk. (2008) Generation of Sierpinski
Triangles: A Case Study for Graph Transformation Tools. LNCS, Springer,
Berlin, Heidelberg, 514-539
2007
B. Bohnet. (2007) IS-FBN, IS-FBS, IS-IAC: The Adaptation of
Two Classic Algorithms for the Generation of Referring Expressions in
order to Produce Expressions like Humans Do. Language Generation and
Machine Translation (UCNLG+MT) workshop at MT Summit XI. Copenhagen, 84-86.
B. Bohnet. (2007) The Induction and Evaluation of Word Order
Rules Using Corpora Based on the Two Concepts of Topological Models.
Language Generation and Machine Translation (UCNLG+MT) workshop at MT
Summit XI. Copenhagen, 38-45.
B. Bohnet. (2007) Covering the Mapping from Semantics up to Morphology by Transducers.
Third International Conference on Meaning-Text Theory. Wiener Slawistischer Almanach, Sonderband 69. 10 pages
B. Bohnet, F. Lareau, L. Wanner. (2007) Automatic Production
of Multilingual Environmental Information. EnviroInfo 2007. Warsaw,
Polen.
Karppinen, A.; Sofiev, M.; Kukkonen, J.;Ranta, H.; Linkosalo, T.;
Nicklaß, D.; Bohnet, B.; Wanner, L. (2007): Pollen: A Challenge for
environmental information services. In: Proceedings of EnviroInfo 2007,
Warsaw.
L. Wanner, B. Bohnet, N. Bouayad-Agha, F. Lareau, A. Lohmeyer,
D. Nicklaß. (2007). On the Challenge of Creating and Communicating Air
Quality Information. In: Procedings of ISESS conference 2007, Prague.
L. Wanner, D. Nicklaß, B. Bohnet, N. Bouayad-Agha, J. Bronder,
F. Ferreira, R. Friedrich, A. Karppinen, F. Lareau, A. Lohmeyer, A.
Panighi, S. Parisio, H. Scheu-Hachtel, J. Serpa. (2007). From
Measurement Data to Environmental Information: MARQUIS - A multimodal
air quality information service for the general public. In: Procedings
of ISESS conference 2007, Prague.
L. Wanner and B. Bohnet. (2007) Automatic Recognition of Lexical
Function Instances. Selected Lexical and Grammatical Issues in the
Meaning-Text Theory. In honour of Igor Mel'cuk. Ed. Leo
Wanner. Studies in Language Companion Series 84. John Benjamins. Pages 231-262.
2006
L. Wanner, B. Bohnet, and M. Giereth. (2006) Making Sense of
Collocations. Computer Speech & Language Volume 20, Issue 4 , October 2006,
Pages 609-624.
Leo Wanner, Bernd Bohnet, and Mark Giereth. (2006) What is beyond Collocations?
Insights from Machine Learning Experiments. EURALEX.
L. Wanner, D. Nicklas, R. Friedrich, A. Panighi, B.
Bohnet, E. Lehmann, S. Parisio, H. Scheu-Hachtel, A.
Lohmeyer, M. Remedio, A. Karppinen, F. Ferreira, J.
Bronder. (2006). Das mltimodale Luft-Qualitdts-Informationssystem MARQUIS. In: IfGI prints 23, Geosensornetzwerke von der Forschung zur praktischen Anwendung, 73-84
2005
Leo Wanner & Bernd Bohnet & Margarita Alonso. (2005) The True, Deep
Happiness: Towards the Automatic Semantic Classification of Adjective-Noun
Collocations. In Kiefer F. et al. (eds.), Papers in Computacional Lexicography (COMPLEX) 2005, Linguistics Institute and Hungarian Academy of Sciences, Budapest, Hungary, 255-265
L. Wanner, B. Bohnet, M. Giereth and V. Vidal. (2005) The first
steps towards the automatic compilation of specialized collocation
dictionaries. Application-Driven Terminology Engineering. John Benjamins
Publishing Company. Volume 11, Number 1, pp. 143-180 (38)
B. Bohnet and R. Dale. (2005)
Viewing Referring Expression
Generation as a Search Problem. Nineteenth International Joint Conference on
Artifical Intelligence (IJCAI). Edinburgh, 1004-1009
2004
B. Bohnet and H. Seniv. (2004) Mapping Dependency Structures to Phrase Structures
and the Automatic Acquisition of Mapping Rules. LREC. Portugal, Lisboa, 855-858
Stefan Klatt & Bernd Bohnet. (2004) You don't have to think twice if you
carefully tokenize. 1st International Joint Conference on Natural Language
Processing (IJCNLP-04), China, Sanya City, 299-309.
B. Bohnet. (2004) A Graph Grammar Approach to Map between
Dependency Trees and Topological Models. The 1st International Joint Conference on
Natural Language Processing (IJCNLP-04). China, Sanya City, 636-645.
2003
Bernd Bohnet. (2003) Mapping Phrase Structures to Dependency Structures in the Case of
Free Word Order Languages. The First International
Conference on Meaning-Text Theory. Paris, 239-249.
L. Wanner, B. Bohnet and M. Giereth. (2003) Deriving the Communicative Structure
in Applied NLG. Proceedings of the 9th European Natural Language
Generation Workshop at the Annual Meeting of the Association for Computational
Linguistics, Budapest, 111-118.
2002
B. Bohnet, S. Klatt and L. Wanner. (2002) An Approach to Automatic Annotation of
Functional Information to Adjectives with an Application to German.
In Third International Conference on Language Resources and Evaluation (LREC
2002), Workshop: Linguistic Knowledge Acquisition and Representation,
Las Palmas, Canary Islands 2002.
2001
B. Bohnet and L. Wanner. (2001) On Using a Parallel Graph Rewriting
Formalism in Generation. In proceedings of the 8th European Natural
Language Generation Workshop at the Annual Meeting of the Association
for Computational Linguistics, Toulouse, 47-56
B. Bohnet ; L. Wanner ; R. Ebel ; B. Knoerzer ; M. Tauber ;
W. Weiss, H. Scheu-Hachtel. Autotext-UIS. (2001) Automatische
Produktion von Ozonkurzberichten im Umweltinformationssystem
Baden-Württemberg. In: Neue Methoden fuer das Wissensmanagement im
Umweltschutz : 4. Workshop des GI Arbeitskreises Hypermedia im
Umweltschutz und Workshop 3 der GI-Initiative Environmental Markup
Language - 29, 21-32.
2000
B. Bohnet, A. Langjahr and L. Wanner. 2000. A Development Environment for
MTT-Based Sentence Generators. In Proceedings of the SEPLN-2000. Vigo, Spain, 35-36.
L. Wanner, B. Bohnet and C. Gerstenberger. 2000. Generation of Air Pollutant
Information. In Proceedings of the SEPLN-2000. Vigo, Spain, 37-40.
B. Bohnet, A. Langjahr and L. Wanner. (2000) A Development Environment
for an MTT-Based Sentence Generator. Proceedings of the First International
Conference on Natural Language Generation. Mitzpe Ramon, Israel, 260-263
Shared Tasks
Generation Challenges 2011 Surface Realisation Shared Task
Ranked 1st: "Across the metrics and data sets, STUMABA-S is consistently the top-scoring system ..."
CoNLL-2009 Shared Task: Syntactic and Semantic Dependencies in Multiple Languages
Ranked 4th for syntactic and semantic parsing; 2nd for dependency parsing: 1st
for German and English dependency parsing
NIST TAC 2009 Shared Task for Summarization as member of the ICSI team
ICSI/UTD Summarization Systems was top ranked due to different metrics
Generation Challenge 2009: GREC-NEG (Named Entity Generation) and GREC-MSR (Main Subject References)
Ranked 1st (NEG) and 3rd (MSR)
Referring Expression Generation Challenge 2008: Attribute Selection for
Referring Expression Generation, Referring Expression Generation
(Ranked somewhere in the middle)
NLG Challenge on Attribute Selection for Generating Referring Expressions (ASGRE) in 2007
Ranked 1st for similarity score (DICE)
Past Projects
Applied Text Generation: Auto-Text
Auto-Text is a system for automatic generation of air polution reports.
The project was founded by the Ministry of Environment and Traffic of the
province Badenwürttemberg.
The Auto-Text system was developted by Leo Wanner and Bernd Bohnet using
an early release of the MATE system.
Since 2001 the system is in summer from april to september online available.
The system is hosted by the UMEG one of the project partners (AutoText via link Ozon-Kurzberichte).
Automatic semantic classifikation of collocations
The goal of the Projekt with the title Acciones Integradas Hispano-Alemans
2003-D/0232983 was the automatic classifikation of Spanish
collocations. The project was supported by the DAAD.
Applied Text Generation - Multimodal Air Quality Information Service for
General Public: Marquis Marquis is a cross-border information service that
provides user-tailored analysis, interpretation, and multilingual communication
for meteorological and air quality information for selected regions in the
EU. The project is part of eContent program of the EU and it lasts from 2005
until 2006. I have lead the work package Multimodal and Multilingual Information Generation.
Innovative Language Technology: IS
Currently, we start a company for Natural Language Processing which is support by the
Technoloy Transfer Initiative of University of Stuttgart and the state of Baden-Württemberg.