B08: Hierarchical frame induction via probabilistic models
Overview
Publications
Talks & Posters
Events
Ressources
Funding period: 01.07.2015 – 30.06.2019
Project B08 aims at the unsupervised induction of frames for events described by verbs, using techniques from probabilistic frame modeling and from machine learning. The outcome should not only identify predicate-argument relations but also provide a rich characterization of these relations via hierarchically ordered semantic roles. This allows then the induction of an event type hierarchy that will be checked against and provide evidence for the event types assumed in the linguistic CRC projects and in the literature. In a second step, B08 also seeks to learn relations between the frames of verbs that are subject to alternations. This goal will be pursued in a semi-supervised approach.
Head of the project
Prof. Dr. Laura Kallmeyer
Institut für Sprache und Information
Abteilung Computerlinguistik
Universitätsstr. 1
40225 Düsseldorf
Phone: +49 211 81-138 99
Email: kallmeyer@phil.uni-duesseldorf.de
Staff
Esther Seyffarth
Telefon: +49 211 81-03185
E-Mail: seyffarth@phil.hhu.de
Alumni
Dr. Behrang QasemiZadeh
Email: zadeh@phil.hhu.de
Roman Gelzhäuser
Email: Roman.Gelzhaeuser@uni-duesseldorf.de
2020
Regina Stodden & Behrang QasemiZadeh (2020). Do you Feel Certain about your Annotation? A Web-based Semantic Frame Annotation Tool Considering Annotators’ Concerns and Behaviors. In Proceedings of the 12th International Conference on Language Resources and Evaluation (LREC 2020), p.7132-7139. (Print), (bib)
2019
Esther Seyffarth. Modeling the Induced Action Alternation and the Caused-Motion Construction with Tree Adjoining Grammar (TAG) and Semantic Frames. In Proceedings of CSTFRS – Computing Semantics with Types, Frames and Related Structures, IWCS 2019.
Behrang QasemiZadeh, Miriam R. L. Petruck, Regina Stodden, Laura Kallmeyer, and Marie Candito: SemEval-2019 Task 2: Unsupervised Lexical Frame Induction. Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), 2019, pp. 16-30. (Print)
Esther Seyffarth. Identifying Participation of Individual Verbs or VerbNet Classes in the Causative Alternation. In Proceedings of the Society for Computation in Linguistics. (Link)
2018
Regina Stodden, Behrang Qasemizadeh, and Laura Kallmeyer: TRAPAAC – a transition based neural dependency parser for identification of VMWEs. Joint Workshop on Linguistic Annotation, Multiword Expressions and Constructions (LAW-MWE-CxG-2018). (Print)
Ramisch, Carlos, Silvio Ricardo Cordeiro, Agata Savary, Veronika Vincze, Behrang QasemiZadeh, MarieCandito, Verginica Barbu Mititelu, Voula Giouli, Ivelina Stoyanova, Nathan Schneider & Timm Lichte: Edition 1.1 of the PARSEME shared task on automatic identification of Verbal Multiword Expressions. InProceedings of LAW-MWE-CxG 2018, Santa Fe, USA. (Print)
Behzad Naderalvojoud, Behrang Qasemizadeh, Laura Kallmeyer, and Ebru Akcapinar Sezer: Cross-Lingual Approach for Building Multilingual Sentiment Lexicons for Neural Sentiment Analysis. 21st International Conference on Text, Speech and Dialogue. Springer. (Preprint)
Laura Kallmeyer, Behrang Qasemizadeh and Jackie Chi Kit Cheung: Coarse Lexical Frame Acquisition at the Syntax–Semantics Interface Using a Latent-Variable PCFG Model. *SEM 2018 – The Seventh Joint Conference on Lexical and Computational Semantics, Association for Computational Linguistics. (Print)
Savary, Agata, Marie Candito, Verginica Barbu Mititelu, Eduard Bejček, Fabienne Cap, Slavomír Čéplö, Silvio Ricardo Cordeiro, Gülşen Eryiğit, Voula Giouli, Maarten van Gompel, Yaakov HaCohen-Kerner, Jolanta Kovalevskaitė, Simon Krek, Chaya Liebeskind, Johanna Monti, Carla Parra Escartín, Lonneke van der Plas, Behrang QasemiZadeh, Carlos Ramisch, Federico Sangati, Ivelina Stoyanova & Veronika Vincze. 2018. Extended papers from the mwe 2017 workshop: Extended papers from the mwe 2017 workshop chap. The PARSEME multilingual corpus of verbal multiword expressions. Berlin, Germany: Language Science Press. (Preprint)
Maldonado, Alfredo & Behrang QasemiZadeh. 2018. Extended papers from the mwe 2017 workshop: Extended papers from the mwe 2017 workshop chap. Analysis and Insights from the PARSEME Shared Task dataset. Berlin, Germany: Language Science Press. (Preprint)
Esther Seyffarth. Verb Alternations and Their Impact on Frame Induction. In Proceedings of NAACL 2018 Student Research Workshop. Association for Computational Linguistics. (Print)
Gábor, Kata, Davide Buscaldi, Anne-Kathrin Schumann, Behrang QasemiZadeh, Haifa Zargayouna & Thierry Charnois. 2018. Semeval-2018 task 7: Semantic relation extraction and classification in scientific papers. In Proceedings of the 12th international workshop on semantic evaluation (semeval-2018), Association for Computational Linguistics. (Print)
Bladier, Tatiana, Esther Seyffarth, Oliver Hellwig & Wiebke Petersen: AET: Web-based adjective exploration tool for German. In Proceedings of LREC 2018 (Print)
Natalia Klyueva, Behrang Qasemizadeh and Anna Vernerova: Querying Multi-word Expressions Annotation with CQL. 16th International Workshop on Treebanks and Linguistic Theories, 2018. (Print)
2017
Behrang QasemiZadeh and Laura Kallmeyer: HHU at SemEval-2017 Task 2: Fast Hash-Based Embeddings for SemanticWord Similarity Assessment. SemEval, ACL, 2017. (Print)
Behzad Naderalvojoud, Behrang Qasemizadeh and Laura Kallmeyer: Lexicon-Based Deep Learning for Contextual Sentiment Analysis. GermEval 2017: Shared Task on Aspect-based Sentiment in Social Media. (Print Bibtex)
Behrang QasemiZadeh, Laura Kallmeyer and Aurélie Herbelot: Projection Aléatoire Non-Négative pour le Calcul de Word Embedding / Non-Negative Randomized Word Embedding. In Proceedings of TALN 2017. (Preprint Resources Bibtex)
Agata Savary; Carlos Ramisch; Silvio Cordeiro; Federico Sangati; Veronika Vincze; Behrang QasemiZadeh; Marie Candito; Fabienne Cap; Voula Giouli; Ivelina Stoyanova; Antoine Doucet: The PARSEME Shared Task on Automatic Identification of Verbal Multiword Expressions, In Proceedings of Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017), 2017, pp. 31-47. Print
2016
Behrang QasemiZadeh: A Study on the Interplay Between the Corpus Size and Parameters of a Distributional Model for Term Classification, In Proceedings of Computerm 2016, pp. 62-72. Print
Behrang QasemiZadeh and Laura Kallmeyer: Random Positive-Only Projections: PPMI-Enabled Incremental Semantic Space Construction, In Proceedings of *Sem 2016: The Fifth Joint Conference On Lexical and Computational Semantics, pp. 189-198. Preprintr; SLIDES
Aurelie Herbelot and Behrang QasemiZadeh: You and me… in a vector space: modelling individual speakers with distributional semantics, In Proceedings of *Sem 2016: The Fifth Joint Conference On Lexical and Computational Semantics, pp. 179-188. Preprint
Behrang QasemiZadeh and Anne Schumann (2016): The ACL RD-TEC 2.0: A Language Resource for Evaluating Term Extraction and Entity Recognition Methods, Proceedings of 10th edition of the Language Resources and Evaluation Conference (LREC 2016). Preprint
2018
Esther Seyffarth: From Text to Frame Generation. NLP Reading Group at CUNY Graduate Center, City University of New York, USA. [Slides]
Esther Seyffarth (2018). Verb Alternations and Their Impact on Frame Induction. In Proceedings of the NAACL Student Research Workshop, Association for Computational Linguistics. [Poster] [Paper]
2017
Laura Kallmeyer and Behrang QasemiZadeh: Unsupervised Frame Induction Through PCFGs. CL Colloquium, Heinrich-Heine-Universität Düsseldorf, 8 June 2017, Düsseldorf, Germany. ( Slides)
Laura Kallmeyer and Rainer Osswald: Polysemy, Coercion and Quantification. CL Colloquium, Heinrich-Heine-Universität Düsseldorf, 1 June 2017, Düsseldorf, Germany.
2016
Behrang QasemiZadeh, Laura Kallmeyer, and Aurelie Herbelot: PPP-Hashing: A Semantic-Sensitive Hashing Technique. University of Cambridge, 10 November 2016, Cambridge, England. (Slides)
Behrang QasemiZadeh: Annotation of Multiword Expressions in the Farsi Section of the Universal Dependencies Project. Relieving the pain in the neck in natural language processing: 7th final general meeting, 26-27 September 2016, Dubrovnik, Croatia. (Abstract; Poster)
Laura Kallmeyer and Behrang QasemiZadeh: The PoP Approach to Vector Semantics. Talk at the Semantics and Pragmatics Exchange (SemPrE) Colloquium at Heinrich-Heine-Universität Düsseldorf, 15 June 2016, Düsseldorf, Germany. (Slides)
24.05.2018 | Invited talk by Paul Buitelaar (NUI Galway, Insight Centre for Data Analytics): CoreLex : Recap and Related Work (slides) |
14.12.2017 | Invited talk by Marie Candito (Université Paris Diderot): From deep syntax to semantic frames: feedback from the French FrameNet project (slides) |
01.06.2017 | Behzad Naderalvojoud (Hacettepe University) visits SFB991 as a visiting fellow with B08 project. |
11.-22.09.2017 | The 9th DGfS-CL Fall School 2017 will take place from 11th September to 22nd September 2017 at Heinrich Heine University Düsseldorf, Germany. |
02.03.2017 | Invited Talk by Jackie C. K. Cheung (McGill University): Generalized Natural Language Generation (slides) |
27.02. – 03.03.2017 | Jackie C. K. Cheung, McGill School of Computer Science, visits SFB991. |
01.02.2017 | Esther Seyffarth joins B08 as a researcher. |
15.12.2016 | Laura Kallmeyer is awarded the prestigious ERC Grant for the research project TreeGraSP (tree rewriting grammar and the syntax-semantics interface: From grammar development to semantic parsing). |
05.10.2016 | Invited Talk by Daniel Peterson (Institute of Cognitive Science, University of Colorado Boulder): Bayesian Semantics, verb sense induction (slides), Links to the induced frames from the Gigaword corpus, the Web corpus) |
26.-27.092016 | Participation in the 7th general meeting of PARSEME (together with A02) |
15.-17.09.2016 | The conference Cognitive Structures:Linguistic, Philosophical and Psychological Perspectives was held by SFB 991 at HHUD. |
17.08.2016 | Invited Talk by Miriam R L Petruck (UC-Berkeley, FrameNet team): FrameNet: Current Status, Recent Activity, and Future Plans (slides) |
114.07.2016 | Invited Talk by Aurélie Herbelot (University of Trento): Where do models come from and where do they go? Underspecifying set-theoretic semantics with vector spaces (slides) |
29.06. – 01.07.2016 | TAG+12 Conference (together with A02, B01, and D04) |
17.-20.05.2016 | Participating in the ENeL Training Event (Tools and methods for creating innovative e-dictionaries), Ljubljana, Slovenia. |
Resources related to the Positive-Only Projection Method for Word Embedding.
The NoSketchEngine for Browsing Corpora at HHU.
The FLAT Server for the Annotation of Multi-Word Expressions (used in the PARSEME shared task.)
Lexical Semantic Frame Annotation System (used for annotation of frames in context)