Elamur Research⌕
☰
home / papers / 2015 / Joint prediction in MST-style discourse parsing for argumentation mining

Joint prediction in MST-style discourse parsing for argumentation mining

Summary

Peldszus and Stede study argumentation mining for short texts. Given a text already split into argumentative discourse units, the goal is to recover its full argument structure: which segment is the central claim, which segments attach to which, whether each segment belongs to the proponent or the opponent, and whether each link supports or attacks its target. They use the parallel German/English arg-microtext corpus and simplify its Freeman-based annotation into trees with only support and attack relations.

Separate classifiers handle four levels: attachment, central claim, role and function. Their probabilities are merged into edge scores of a single "evidence graph," and a standard maximum spanning tree (MST) algorithm then picks the best global structure. This joint decoding beats simple baselines and two learned models without global optimisation on relation identification. It also improves central claim and function prediction and does well against a multi-pass mstparser pipeline.

The authors present this as the first data-driven argument parser that optimises structure over the whole text. It is also the first to treat segment typing and relation prediction together.

Key points

  • Data: the arg-microtext corpus contains 112 texts with 576 ADUs, about 5 segments per text, each with one explicit central claim. It was written in German and professionally translated into English.
  • Imbalance and distance: 464 segment pairs carry a relation and roughly 2000 pairs do not, so attachment classes are heavily skewed. 46% of relations link non-adjacent segments.
  • Baseline: the first-segment baseline (BL-first) exploits the fact that 37.9% of relations point to the first segment.
  • Limitation: role is the only level where MST decoding does not beat the base classifier.
  • Simulations: the authors artificially improved individual base classifiers and measured the effect in κ. Better function classification helps every level, including attachment. Combining edge scores as a weighted sum versus a product of probabilities gave no significant difference.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics: Discourse analysis

Cited by

82 papers of our library cite it (of the 140673 whose bibliographies we read); 4 of them have a page here, 78 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/d15-1110

@inproceedings{peldszus2015joint,
  title = {{Joint prediction in MST-style discourse parsing for argumentation mining}},
  author = {Peldszus, Andreas and Stede, Manfred},
  booktitle = {Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing},
  year = {2015},
  doi = {10.18653/v1/d15-1110},
  url = {https://doi.org/10.18653/v1/d15-1110}
}

References

27 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Jason Baldridge, Nicholas Asher, and Julie Hunter. 2007. Annotation for and robust parsing of discourse structure on unrestricted texts. Zeitschrift für Sprachwissenschaft, 26:213–239. find
  2. Bernd Bohnet. 2010. Very high accuracy and fast dependency parsing is not a contradiction. In Proceedings of the 23rd International Conference on Computational Linguistics, COLING ’10, pages 89–97, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
  3. Jill Burstein and Daniel Marcu. 2003. A machine learning approach for identification thesis and conclusion statements in student essays. Computers and the Humanities, 37(4):455–467. find
  4. Y. J. Chu and T. H. Liu. 1965. On the shortest arborescence of a directed graph. Science Sinica, 14:1396– 1400. find
  5. Jack Edmonds. 1967. Optimum Branchings. Journal of Research of the National Bureau of Standards, 71B:233–240. find
  6. Vanessa Wei Feng and Graeme Hirst. 2011. Classifying arguments by scheme. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1, HLT ’11, pages 987–996, Stroudsburg, PA, USA. Association for Computational Linguistics. find
  7. James B. Freeman. 1991. Dialectics and the Macrostructure of Argument. Foris, Berlin. find
  8. James B. Freeman. 2011. Argument Structure: Representation and Theory. Argumentation Library (18). Springer. find
  9. Saidul Kazi Hasan and Vincent Ng. 2013. Stance classification of ideological debates: Data, models, features, and constraints. In Proceedings of the Sixth International Joint Conference on Natural Language Processing, pages 1348–1356. Asian Federation of Natural Language Processing. find
  10. John Lawrence, Chris Reed, Colin Allen, Simon McAlister, and Andrew Ravenscroft. 2014. Mining arguments from 19th century philosophical texts using topic based modelling. In Proceedings of the First Workshop on Argumentation Mining, pages 79–87, Baltimore, Maryland, June. Association for Computation… find
  11. Maria Liakata, Shyamasree Saha, Simon Dobnik, Colin R. Batchelor, and Dietrich Rebholz-Schuhmann. 2012. Automatic recognition of conceptualization zones in scientific articles and two life science applications. Bioinformatics, 28(7):991– 1000. find
  12. Ryan McDonald, Koby Crammer, and Fernando Pereira. 2005a. Online large-margin training of dependency parsers. In Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL’05), pages 91–98, Ann Arbor, Michigan, June. Association for Computational Linguistics. find our card
  13. Ryan McDonald, Fernando Pereira, Kiril Ribarov, and Jan Hajic. 2005b. Non-projective dependency parsing using spanning tree algorithms. In Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, pages 523–530, Vancouver, British Columbi… find our card
  14. Philippe Muller, Stergos Afantenos, Pascal Denis, and Nicholas Asher. 2012. Constrained decoding for text-level discourse parsing. In Proceedings of COLING 2012, pages 1883–1900, Mumbai, India, December. find our card
  15. The COLING 2012 Organizing Committee. find
  16. Raquel Mochales Palau and Marie-Francine Moens. 2011. Argumentation mining. Artificial Intelligence and Law, 19(1):1–22. find
  17. Joonsuk Park and Claire Cardie. 2014. Identifying appropriate support for propositions in online user comments. In Proceedings of the First Workshop on Argumentation Mining, pages 29–38, Baltimore, Maryland, June. Association for Computational Linguistics. find
  18. F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. 2011. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:282… find
  19. Andreas Peldszus and Manfred Stede. 2013. From argument diagrams to automatic argument mining: A survey. International Journal of Cognitive Informatics and Natural Intelligence (IJCINI), 7(1):1–31. find
  20. Andreas Peldszus and Manfred Stede. to appear. An annotated corpus of argumentative microtexts. In Proceedings of the First European Conference on Argumentation: Argumentation and Reasoned Action, Lisbon, Portugal, June 2015. find
  21. Andreas Peldszus. 2014. Towards segment-based recognition of argumentation structure in short texts. In Proceedings of the First Workshop on Argumentation Mining, Baltimore, U.S., June. Association for Computational Linguistics. find
  22. Christian Stab and Iryna Gurevych. 2014. Identifying argumentative discourse structures in persuasive essays. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 46–56, Doha, Qatar, October. Association for Computational Linguistics. find our card
  23. Manfred Stede. 2002. DiMLex: A Lexical Approach to Discourse Markers. In Vittorio Di Tomaso Alessandro Lenci, editor, Exploring the Lexicon - Theory and Computation. Edizioni dell’Orso, Alessandria, Italy. find
  24. Simone Teufel and Min-Yen Kan. 2011. Robust argumentative zoning for sensemaking in scholarly documents. In Raffaella Bernadi, Sally Chambers, Björn Gottfried, Frédérique Segond, and Ilya Zaihrayeu, editors, Advanced Language Technologies for Digital Libraries, volume 6699 of Lecture Notes in Com… find
  25. Simone Teufel and Marc Moens. 2002. Summarizing scientific articles: Experiments with relevance and rhetorical status. Comput. Linguist., 28(4):409–445, December. find our card
  26. Simone Teufel. 2010. The Structure of Scientific Articles: Applications to Citation Indexing and Summarization. CSLI Studies in Computational Linguistics. CSLI Publications. find
  27. Douglas Walton, Chris Reed, and Fabrizio Macagno. 2008. Argumentation Schemes. Cambridge University Press. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/peldszus-2015-joint-prediction-mst-style-discourse/