Elamur Research⌕
☰
home / papers / 2020 / GerDraCor-Coref: A Coreference Corpus for Dramatic Texts in German

GerDraCor-Coref: A Coreference Corpus for Dramatic Texts in German

Summary

Computational studies of drama have mostly relied on structural features such as character co-presence networks. These miss how characters are portrayed in other characters' speech, and they ignore non-speaking persons and plot-relevant objects. Coreference resolution (CR) could address this, but no German drama data existed for building or testing such systems. Pagel and Reiter present GerDraCor-Coref, a manually annotated coreference corpus built on top of the TEI/XML-encoded GerDraCor collection. It covers 31 German plays written between 1730 and 1920, mostly single acts, with a few plays annotated in full. Eight trained literature students did the annotation following guidelines derived from TüBa-D/Z. A quantitative comparison with German news and radio corpora shows that the reference structure of plays differs strongly from news text and is closer to dialogical material such as interviews. Mentions in the plays are denser and shorter. As a baseline, the rule-based CorZu system reaches only 28.8 CoNLL. This poor result shows that existing resolvers transfer badly to drama, and it makes the corpus a useful benchmark for domain adaptation in computational literary studies.

Key points

  • Data and coverage: The corpus contains 31 plays with about 298k tokens and about 61k annotated mentions. Each decade between 1730 and 1920 is represented by at least one play.
  • Formats and availability: The corpus is distributed on GitHub (quadrama/gerdracor-coref) and has a Zenodo DOI. It comes in three formats:
    • CoNLL, with an added speaker column in the style of GRAIN;
    • XMI, for Apache UIMA;
    • TEI/XML.
  • Annotation quality: Annotators used the CorefAnnotator tool. Inter-annotator agreement was 62.3 CoNLL (78.7 MUC F), somewhat below the 83 MUC reported for TüBa-D/Z.
  • Comparison with other corpora: Mention density is 0.20, compared with 0.09 for TüBa-D/Z. Average mention length is 1.52 tokens, compared with 2.10. Documents are much longer, averaging about 6,630 tokens versus about 467 in TüBa-D/Z.
  • Drama-specific difficulties: The authors report problems with disguises and character mix-ups, characters whose names change, vaguely defined groups, and very long coreference chains. They plan to use dramatis personae information in future CR models.

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: Coreference resolution

Cited by

6 papers of our library cite it (of the 140673 whose bibliographies we read); 2 of them have a page here, 4 not yet.

Cite

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

@inproceedings{pagel2020gerdracor,
  title = {{GerDraCor-Coref: A Coreference Corpus for Dramatic Texts in German}},
  author = {Pagel, Janis and Reiter, Nils},
  booktitle = {Proceedings of the Twelfth Language Resources and Evaluation Conference},
  year = {2020},
  doi = {10.5281/zenodo.4737163},
  url = {https://doi.org/10.5281/zenodo.4737163}
}

References

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

  1. Bamman, D., Underwood, T., and Smith, N. A. (2014). A bayesian mixed effects model of literary character. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (ACL), pages 370– 379. find our card
  2. Bamman, D., Popat, S., and Shen, S. (2019). An annotated dataset of literary entities. In Proceedings of NAACL. find
  3. Baumann, S. and Riester, A. (2012). Referential and lexical givenness: Semantic, prosodic and cognitive aspects. In Gorka Elordieta et al., editors, Prosody and Meaning, volume 25 of Interface of Explorations, pages 119–162. Mouton de Gruyter, Berlin. ↗
  4. Björkelund, A. and Kuhn, J. (2014). Learning structured perceptrons for coreference resolution with latent antecedents and non-local features. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (ACL), pages 47–57. find our card
  5. Björkelund, A., Eckart, K., Riester, A., Schauffler, N., and Schweitzer, K. (2014). The extended DIRNDL corpus as a resource for automatic coreference and bridging resolution. In Proceedings of the 9th International Conference on Language Resources and Evaluation (LREC), pages 3222–3228. find
  6. Bullard, J. and Ovesdotter Alm, C. (2014). Computational analysis to explore authors’ depiction of characters. In Proceedings of the 3rd Workshop on Computational Linguistics for Literature (CLFL), pages 11–16, Gothenburg, Sweden. Association for Computational Linguistics. find
  7. Clark, K. and Manning, C. D. (2016). Improving coreference resolution by learning entity-level distributed representations. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (ACL), pages 643–653. find our card
  8. Davis, P. T., Elson, D. K., and Klavans, J. L. (2003). Methods for precise named entity matching in digital collections. In Proceedings of the 3rd ACM/IEEE-CS Joint Conference on Digital Libraries, JCDL ’03, pages 125– 127, Washington, DC, USA. IEEE Computer Society. ↗
  9. Hamp, B. and Feldweg, H. (1997). GermaNet - a lexical-semantic net for German. In Proceedings of the ACL workshop Automatic Information Extraction and Building of Lexical Semantic Resources for NLP Applications, pages 9–15. find our card
  10. Kolhatkar, V., Roussel, A., Dipper, S., and Zinsmeister, H. (2018). Anaphora with non-nominal antecedents in computational linguistics: A survey. Computational Linguistics, 44(3):547–612. ↗ our card
  11. Krug, M., Puppe, F., Jannidis, F., Macharowsky, L., Reger, I., and Weimer, L. (2015). Rule-based coreference resolution in german historic novels. In Proceedings of the Fourth Workshop on Computational Linguistics for Literature, pages 98–104. find our card
  12. Krug, M., Puppe, F., Reger, I., Weimer, L., Macharowsky, L., Feldhaus, S., and Jannidis, F. (2018). Description of a Corpus of Character References in German Novels – DROC [Deutsches ROman Corpus]. DARIAH-DE Working Papers, 27. find
  13. Martschat, S. (2017). Structured Representations for Coreference Resolution. Ph.D. thesis, Heidelberg University. ↗
  14. Naumann, K., (2007). Manual for the Annotation of in-document Referential Relations. Abt. Computerlinguistik Universität Tübingen. find
  15. Pradhan, S., Moschitti, A., Xue, N., Uryupina, O., and Zhang, Y. (2012). CoNLL-2012 shared task: Modeling multilingual unrestricted coreference in OntoNotes. In Joint Conference on EMNLP and CoNLL - Shared Task, CoNLL 2012, pages 1–40, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
  16. Pradhan, S., Luo, X., Recasens, M., Hovy, E., Ng, V., and Strube, M. (2014). Scoring coreference partitions of predicted mentions: A reference implementation. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (ACL), pages 30–35. find our card
  17. Raghunathan, K., Lee, H., Rangarajan, S., Chambers, N., Surdeanu, M., Jurafsky, D., and Manning, C. (2010). A multi-pass sieve for coreference resolution. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 492– 501. find our card
  18. Reiter, N., Blessing, A., Echelmeyer, N., Kremer, G., Koch, S., Murr, S., Overbeck, M., and Pichler, A. (2017). CUTE: CRETA Unshared Task zu Entitätenreferenzen. In Proceedings of DHd. find
  19. Reiter, N. (2018). CorefAnnotator - A New Annotation Tool for Entity References. In Abstracts of EADH: Data in the Digital Humanities. ↗
  20. Riester, A. and Baumann, S. (2017). The RefLex Scheme – Annotation guidelines. SinSpeC. Working papers of the SFB 732 Vol. 14, University of Stuttgart. ↗
  21. Rösiger, I. and Kuhn, J. (2016). IMS HotCoref DE: A data-driven co-reference resolver for German. In Proceedings of the 10th International Conference on Language Resources and Evaluation (LREC), pages 155–160. find our card
  22. Rösiger, I., Schulz, S., and Reiter, N. (2018). Towards coreference for literary text: Analyzing domain-specific phenomena. In Proceedings of the Second Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature, pages 129–138. find our card
  23. Schweitzer, K., Eckart, K., Gärtner, M., Faleńska, A., Riester, A., Rösiger, I., Schweitzer, A., Stehwien, S., and Kuhn, J. (2018). German radio interviews: The GRAIN release of the SFB732 Silver Standard Collection. In Proceedings of the 11th International Conference on Language Resources and Evaluation, LREC 2018. find
  24. Telljohann, H., Hinrichs, E., and Kübler, S. (2004). The Tüba-D/Z treebank: Annotating German with a context-free backbone. In Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC), pages 2229–2232. find
  25. Toutanova, K., Klein, D., Manning, C., and Singer, Y. (2003). Feature-rich part-of-speech tagging with a cyclic dependency network. In Proccedings of the Human Language Technology Conference and the North American Chapter of the Association for Computational Linguistics (HLT-NAACL), pages 173–180. find
  26. Trilcke, P., Fischer, F., and Kampkaspar, D. (2015). Digital network analysis of dramatic texts. In DH2015 Conference Abstracts. ↗
  27. Tuggener, D. (2016). Incremental Coreference Resolution for German. Ph.D. thesis, University of Zürich. ↗
  28. Versley, Y. (2006). Disagreement dissected: Vagueness as a source of ambiguity in nominal (co-)reference. In ESS-LLI 2006 Workshop on Ambiguity in Anaphora, pages 83–89. find
  29. Vishnubholta, K., Hammond, A., and Hirst, G. (2019). Are fictional voices distinguishable? classifying character voices in modern drama. In Proceedings of the 3rd Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature, pages 29–34. find
  30. Yang, J. B., Mao, Q., Xiang, Q. L., Tsang, I. W., Chai, K. M. A., and Chieu, H. L. (2012). Domain adaptation for coreference resolution: An adaptive ensemble approach. In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Comutational Natural Language Learning (EMNLP-CoNLL), pages 744–753. find our card
  31. Zhao, S. and Ng, H. T. (2014). Domain adaptation with active learning for coreference resolution. In Proceedings of the 5th International Workshop on Health Mining and Information Analysis (Louhi), pages 21–29. 10. find our card
  32. Language Resource References Bamman, D., Popat, S., and Shen, S. (2019). An annotated dataset of literary entities. In Proceedings of NAACL. find
  33. Björkelund, A., Eckart, K., Riester, A., Schauffler, N., and Schweitzer, K. (2014). The extended DIRNDL corpus as a resource for automatic coreference and bridging resolution. In Proceedings of the 9th International Conference on Language Resources and Evaluation (LREC), pages 3222–3228. find
  34. Krug, M., Puppe, F., Reger, I., Weimer, L., Macharowsky, L., Feldhaus, S., and Jannidis, F. (2018). Description of a Corpus of Character References in German Novels – DROC [Deutsches ROman Corpus]. DARIAH-DE Working Papers, 27. find
  35. Reiter, N., Blessing, A., Echelmeyer, N., Kremer, G., Koch, S., Murr, S., Overbeck, M., and Pichler, A. (2017). CUTE: CRETA Unshared Task zu Entitätenreferenzen. In Proceedings of DHd. find
  36. Schweitzer, K., Eckart, K., Gärtner, M., Faleńska, A., Riester, A., Rösiger, I., Schweitzer, A., Stehwien, S., and Kuhn, J. (2018). German radio interviews: The GRAIN release of the SFB732 Silver Standard Collection. In Proceedings of the 11th International Conference on Language Resources and Evaluation, LREC 2018. find
  37. Telljohann, H., Hinrichs, E., and Kübler, S. (2004). The Tüba-D/Z treebank: Annotating German with a context-free backbone. In Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC), pages 2229–2232. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/pagel-2020-gerdracor-coref-coreference-corpus-dramatic/