Cover not available
Part of
The Swedish FrameNet++: Harmonization, integration, method development and practical language technology applications
Edited by Dana Dannélls, Lars Borin and Karin Friberg Heppin
[Natural Language Processing 14] 2021
► pp. 263–280

Chapter 10
Semantic role labeling

| University of Gothenburg
| HeppiLing AB
| University of Gothenburg

We investigate the feasibility of automatic semantic role labeling (SRL) using Swedish FrameNet (SweFN). In the first part of the chapter, we describe a baseline system using a traditional division into segmentation and labeling steps. These subsystems are implemented as separate machine learning models, and we explore a wide range of syntactic and lexical features for these models. In the second part, we turn to the question of how the frame-to-frame relations defined in FrameNet allow us to use the annotated examples more effectively. The cross-frame generalization methods reduce the number of errors made by the labeling classifier by 27%. For previously unseen frames, the reduction is even more significant: 50%.

Article outline
  • 1.Introduction
  • 2.The Swedish FrameNet
  • 3.Semantic role labeling with SweFN
    • 3.1Segmentation and labeling classifiers
  • 4.Experiments
    • 4.1Experimental data and preprocessing
    • 4.2Cross-validation over sentences
    • 4.3Cross-frame role label generalization
    • 4.4Analysis of features
    • 4.5Cross-validation over frames
    • 4.6Increasing classifier robustness by adding cluster features
    • 4.7The effect of syntactic parser choice
    • 4.8Evaluation in the medical domain
    • 4.9Summary of results for the baseline systems
  • 5.Using the FrameNet relational structure to improve the semantic role labeler
    • 5.1A classifier using non-atomic semantic role labels
    • 5.2Generalization methods
  • 6.Experiments in cross-frame generalization
  • 7.Conclusion
  • Notes
  • References
Available under the Creative Commons Attribution-NonCommercial-NoDerivatives (CC BY-NC-ND) 4.0 license.

For any use beyond this license, please contact the publisher at [email protected].

Published online: 26 November 2021
DOI logo
https://doi.org/10.1075/nlp.14.10joh
References
Baker, Collin F., Charles J. Fillmore & John B. Lowe
1998 The Berkeley FrameNet project. In Proceedings of Coling 1998, 86–90. Montréal: ACL.Google Scholar
Borin, Lars, Dana Dannélls, Markus Forsberg, Maria Toporowska Gronostaj & Dimitrios Kokkinakis
2010 The past meets the present in Swedish FrameNet++. In Proceedings of EURALEX 2010, 269–281. Ljouwert/ Leeuwarden: Fryske Akademy.Google Scholar
Borin, Lars, Markus Forsberg & Lennart Lönngren
2013 SALDO: A touch of yin to WordNet’s yang. Language Resources and Evaluation 47(4): 1191–1211. DOI logoGoogle Scholar
Brown, Peter F., Peter V. deSouza, Robert L. Mercer, Vincent J. Della Pietra & Jenifer C. Lai
1992 Class-based n-gram models of natural language. Computational Linguistics 18(4): 467–479.Google Scholar
Collobert, Ronan & Jason Weston
2007 Fast semantic extraction using a novel neural network architecture. In Proceedings of ACL 2007, 560–567. Prague: ACL.Google Scholar
Crammer, Koby, Ofer Dekel, Joseph Keshet, Shai Shalev-Schwartz & Yoram Singer
2006 Online passive-aggressive algorithms. Journal of Machine Learning Research 7: 551–585.Google Scholar
Das, Dipanjan, Nathan Schneider, Desai Chen & Noah A. Smith
2010 Probabilistic frame-semantic parsing. In Proceedings of NAACL/HLT 2010, 948–956. Los Angeles.Google Scholar
Dowty, David R.
1991 Thematic proto-roles and argument selections. Language 67(3): 574–619. DOI logoGoogle Scholar
Ejerhed, Eva, Gunnel Källgren, Ola Wennstedt & Magnus Åström
1992 The linguistic annotation system of the Stockholm–Umeå corpus project – description and guidelines. Research report. Umeå: Umeå University, Dept. of Linguistics.Google Scholar
Fan, Rong-En, Kai-Wei Chang, Cho-Jui Hsieh, Xiang-Rui Wang & Chih-Jen Lin
2008 Liblinear: A library for large linear classification. Journal of Machine Learning Research 9: 1871–1874.Google Scholar
Fillmore, Charles J. & Collin Baker
2009 A frames approach to semantic analysis. In Bernd Heine & Heiko Narrog (eds.), The Oxford handbook of linguistic analysis, 313–339. Oxford: Oxford University Press.Google Scholar
Friberg Heppin, Karin & Maria Toporowska Gronostaj
2012 The rocky road towards a Swedish FrameNet – creating SweFN. In Proceedings of LREC 2012, 256–261. Istanbul: ELRA.Google Scholar
Fürstenau, Hagen & Mirella Lapata
2009 Semi-supervised semantic role labeling. In Proceedings of EACL 2009, 220–228. Athens. DOI logoGoogle Scholar
Gildea, Daniel & Daniel Jurafsky
2002 Automatic labeling of semantic roles. Computational Linguistics 28(3): 245–288. DOI logoGoogle Scholar
Gildea, Daniel & Martha Palmer
2002 The necessity of syntactic parsing for predicate argument recognition. In Proceedings of ACL 2002, 239–246. Philadelphia: ACL.Google Scholar
Halácsy, Péter, András Kornai & Csaba Oravecz
2007 HunPos – an open source trigram tagger. In Proceedings of ACL 2007: Demo and poster sessions, 209–212. Prague: ACL.Google Scholar
Johansson, Richard
2008 Dependency-based semantic analysis of natural-language text. Lund University. (PhD thesis).
2009 Statistical bistratal dependency parsing. In Proceedings of EMNLP 2009, 561–569. Singapore: ACL. DOI logoGoogle Scholar
2012 Non-atomic classification to improve a semantic role labeler for a low-resource language. In Proceedings of *SEM 2012, 95–99. Montréal: ACL.Google Scholar
Johansson, Richard, Karin Friberg Heppin & Dimitrios Kokkinakis
2012 Semantic role labeling with the Swedish FrameNet. In Proceedings of LREC 2012, 3697–3700. Istanbul: ELRA.Google Scholar
Johansson, Richard & Pierre Nugues
2006 A FrameNet-based semantic role labeler for Swedish. In Proceedings of Coling/ACL 2006, 436–443. Sydney: ACL. DOI logoGoogle Scholar
2007a LTH: Semantic structure extraction using nonprojective dependency trees. In Proceedings of SemEval 2007, 227–230. Prague: ACL. DOI logoGoogle Scholar
2007b Using WordNet to extend FrameNet coverage. In Proceedings of the Nodalida workshop FRAME 2007: Building frame semantics resources for Scandinavian and Baltic languages, 27–30. Lund: Lund University.Google Scholar
2008a Dependency-based syntactic–semantic analysis with PropBank and NomBank. In Proceedings of the CoNLL 2008 shared task, 183–187. Manchester: ACL. DOI logoGoogle Scholar
2008b The effect of syntactic representation on semantic role labeling. In Proceedings of Coling 2008, 393–400. Manchester: ACL. DOI logoGoogle Scholar
Koo, Terry, Xavier Carreras & Michael Collins
2008 Simple semi-supervised dependency parsing. In Proceedings of ACL/HLT 2008, 595–603. Columbus: ACL.Google Scholar
Màrquez, Lluís, Xavier Carreras, Ken Litkowski & Suzanne Stevenson
2008 Semantic role labeling: An introduction to the special issue. Computational Linguistics 34(2): 145–159. DOI logoGoogle Scholar
Matsubayashi, Yuichiroh, Naoaki Okazaki & Jun’ichi Tsujii
2009 A comparative study on generalization of semantic roles in framenet. In Proceedings of ACL/IJCNLP 2009, 19–27. Singapore: ACL. DOI logoGoogle Scholar
Moschitti, Alessandro, Qi Ju & Richard Johansson
2012 Modeling topic dependencies in hierarchical text categorization. In Proceedings of ACL 2012, 759–767. Jeju: ACL.Google Scholar
Nilsson, Jens, Johan Hall & Joakim Nivre
2006 MAMBA meets TIGER: Reconstructing a Swedish treebank from antiquity. In Treebanking for discourse and speech: Proceedings of the NODALIDA 2005 special session on treebanks for spoken language and discourse, 119–132. Copenhagen: Samfundslitteratur.Google Scholar
Nivre, Joakim, Johan Hall, Jens Nilsson, Atanas Chanev, Gülşen Eryığıt, Sandra Kübler, Svetoslav Marinov & Erwin Marsi
2007 MaltParser: A languageindependent system for data-driven dependency parsing. Natural Language Engineering 13(2): 95–135. DOI logoGoogle Scholar
Palmer, Martha, Dan Gildea & Paul Kingsbury
2005 The proposition bank: An annotated corpus of semantic roles. Computational Linguistics 31(1): 71–105. DOI logoGoogle Scholar
Palmer, Martha, Daniel Gildea & Nianwen Xue
2010 Semantic role labeling. Synthesis Lectures on Human Language Technologies 3(1): 1–103. DOI logoGoogle Scholar
Roth, Dan & Yuancheng Tu
2009 Aspect guided text categorization with unobserved labels. In Proceedings of ICDM 2009, 962–967. Miami Beach: IEEE. DOI logoGoogle Scholar
Xue, Nianwen & Martha Palmer
2004 Calibrating features for semantic role labeling. In Proceedings of EMNLP 2004, 88–94. Barcelona: ACL.Google Scholar