Mova0.5 pre-release⌕
☰
home / docs / papers / English / Frustratingly Hard Domain Adaptation for Dependency Parsing

Frustratingly Hard Domain Adaptation for Dependency Parsing

What it is about

The authors report their entry to the 2007 CoNLL shared task on domain adaptation, where parsers had to adapt to new domains without labelled target data. Their unlabelled dependency parser, used without any adaptation, placed third of twelve teams, within 1% of the top system. No team improved target-domain accuracy much over that baseline. The error analysis points mainly to differences in annotation guidelines between the treebanks. The result is a warning that adaptation methods cannot repair conventions that differ between the source and the target treebank.

Ideas

For Mova

Touches 08. National varieties and 37. Register and variation. The mixed training set (EWT, ESLSpok, CHILDES, ATIS and the school corpus) is exactly the setting where guideline differences decide the result. Measure I1 first, because a conversion fix costs less than any new model.

Topics: Dependency parsing · Treebanks

Cite

The paper itself, not this card. BibTeX from the library's record of the work.

@inproceedings{dredze2007frustratingly,
  title = {{Frustratingly Hard Domain Adaptation for Dependency Parsing}},
  author = {Dredze, Mark and Blitzer, John and Talukdar, Partha Pratim and Ganchev, Kuzman and Graça, João and Pereira, Fernando},
  booktitle = {Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL)},
  year = {2007},
  url = {https://aclanthology.org/D07-1112/}
}

References

13 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Shai Ben-David, John Blitzer, Koby Crammer, and Fernando Pereira. 2006. Analysis of representations for domain adaptation. In NIPS. find
  2. Leo Breiman. 1996. Bagging predictors. Learning, 24(2):123–140. find our card
  3. Machine R. Brown. 1973. A First Language: The Early Stages. Harvard University Press. find
  4. Koby Crammer, Ofer Dekel, Joseph Keshet, Shai Shalev-Shwartz, and Yoram Singer. 2006. Online passive-aggressive algorithms. Journal of Machine Learning Research, 7:551–585, Mar. find
  5. R. Johansson and P. Nugues. 2007. Extended constituent-to-dependency conversion for English. In Proc. of the 16th Nordic Conference on Computational Linguistics (NODALIDA). find our card
  6. Sandra Kübler. 2006. How do treebank annotation schemes influence parsing results? or how not to compare apples and oranges. In RANLP. find
  7. S. Kulick, A. Bies, M. Liberman, M. Mandel, R. Mc-Donald, M. Palmer, A. Schein, and L. Ungar. 2004. Integrated annotation for biomedical information extraction. In Proc. of the Human Language Technology Conference and the Annual Meeting of the North American Chapter of the Association for Computational Linguistics (HLT/NAACL). find
  8. B. MacWhinney. 2000. The CHILDES Project: Tools for Analyzing Talk. Lawrence Erlbaum. find
  9. M. Marcus, B. Santorini, and M. Marcinkiewicz. 1993. Building a large annotated corpus of English: the Penn Treebank. Computational Linguistics, 19(2):313–330. find our card
  10. Ryan McDonald, Kevin Lerman, and Fernando Pereira. 2006. Multilingual dependency parsing with a two-stage discriminative parser. In Conference on Natural Language Learning (CoNLL). find
  11. J. Nivre, J. Hall, S. Kübler, R. McDonald, J. Nilsson, S. Riedel, and D. Yuret. 2007. find
  12. The CoNLL 2007 shared task on dependency parsing. In Proc. of the CoNLL 2007 Shared Task. Joint Conf. on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL). find
  13. Lawrence Saul and Fernando Pereira. 1997. Aggregate and mixed-order markov models for statistical language modeling. In EMNLP. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2007-dredze-frustratingly-hard-domain-adaptation-dependency/