Elamur Research⌕
☰
home / papers / 2012 / Semi-Automatic Annotation of Intra-Sentential Discourse Relations in PDT

Semi-Automatic Annotation of Intra-Sentential Discourse Relations in PDT

Summary

The paper explains how discourse relations inside sentences, signalled by explicit connectives, were added to the Prague Dependency Treebank 2.5 (PDT). PDT is a Czech journalistic treebank with deep-syntax (tectogrammatical) annotation. Most discourse corpora are built on raw text, but the Czech project places discourse links directly on tectogrammatical trees.

A first, largely manual phase covered inter-sentential relations. It also covered intra-sentential cases where the functor did not reliably indicate the discourse sense. In the second phase, a script went through every tree and converted the remaining clause-level dependencies into discourse relations:

  • Finite verbs with causal, conditional, concessive and similar functors became vertical (subordinate) relations.
  • Coordination nodes joining clauses became horizontal relations.

The script derived discourse type, arrow direction, argument positions and connective from the existing annotation. Temporal clauses were first classified by a human and then processed automatically. A few problematic constructions were corrected by hand afterward.

The work shows that deep-syntax annotation predicts discourse structure well enough to automate most intra-sentential annotation. The resulting resource was released openly.

Key points

  • Every count reported is based on PDT's training section together with its development-test section (43,955 sentences, about 9/10 of the treebank). The evaluation test portion was kept largely unobserved.
  • PDT contains 18,187 annotated discourse relations in total: 12,673 intra-sentential and 5,514 inter-sentential (all manual).
  • 9,991 relations were created fully automatically (2,599 vertical, 7,392 horizontal). Another 491 temporal relations were semi-automatic, and 2,191 intra-sentential ones (510 vertical, 1,681 horizontal) were manual.
  • Functors used for automatic detection:
    • Vertical relations: CAUS, COND, CNCS, AIM, CONTRD, SUBS.
    • Horizontal relations: REAS, CSQ, ADVS, CONFR, GRAD, CONJ, DISJ.
    • Remaining ADVS cases defaulted to opposition.
  • Post-correction involved fixing atypical connectives in 23 cases and manually adding 146 clauses tied to complex predicates. The data were published in 2012 under a Creative Commons licence via LINDAT-Clarin.

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 · Treebanks

Cited by

7 papers of our library cite it (of the 140673 whose bibliographies we read); 3 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{jinova2012semi,
  title = {{Semi-Automatic Annotation of Intra-Sentential Discourse Relations in PDT}},
  author = {Jínová, Pavlína and Mírovský, Jiří and Poláková, Lucie},
  booktitle = {Proceedings of the Workshop on Advances in Discourse Analysis and its Computational Aspects},
  year = {2012},
  url = {https://aclanthology.org/W12-4704/}
}

References

13 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Al-Saif, A; Markert, K. (2010). The Leeds Arabic Discourse Treebank: Annotating discourse connectives for Arabic. In Proceedings of the 7th International Conference on Language Resources and Evaluation (LREC 2010), Valletta, Malta, pp. 2046–2053. find our card
  2. Bejček, E., Panevová, J., Popelka, J., Smejkalová, L., Straňák, P., Ševčíková, M., Štěpánek, J., Toman, J., Žabokrtský, Z., Hajič, J. (2012). Prague Dependency Treebank 2.5. Data/software, Charles University in Prague, Czech Republic, http://ufal.mff.cuni.cz/pdt2.5/. ↗
  3. Carlson, L., Marcu, D., and Okurowski, M.E. (2002). RST Discourse Treebank, LDC2002T07 [Corpus]. Linguistic Data Consortium, Philadelphia, PA, USA. find
  4. Hajič, J., Panevová, J., Hajičová, E., Sgall, P., Pajas, P., Štěpánek, J., Havelka, J., Mikulová, M., Žabokrtský, Z. and Ševčíková-Razímová, M. (2006). Prague Dependency Treebank 2.0. Software prototype, Linguistic Data Consortium, Philadelphia, PA, USA, ISBN 1-58563-370-4, www.ldc.upenn.edu, Jul 20… find
  5. Mírovský, J., Jínová, P., Poláková, L. (2012). Does Tectogrammatics help the Annotation of Discourse? In Procedings of the 24th International Conference on Computational Linguistics (COLING 2012), Mumbai, India, December 2012. find
  6. Jínová, P., Mírovský, J., Poláková, L. (2012). Analyzing the Most Common Errors in the Discourse Annotation of the Prague Dependency Treebank. In Proceedings of the 11th International Workshop on Treebanks and Linguistic Theories (TLT 11), Lisboa, Portugal. find
  7. Mikulová, M., Bémová, A., Hajič, J., Hajičová, E., Havelka, J., Kolářová-Řezníčková, V., Kučová, L., Lopatková, M., Pajas, P., Panevová, J., Razímová, M., Sgall, P., Štěpánek, J., Urešová, Z., Veselá, K. and Žabokrtský, Z. (2005). Anotace na tektogramatické rovině Pražského závislostního korpusu. An… find
  8. Mírovský, J., Mladová, L., Žabokrtský, Z. (2010). Annotation Tool for Discourse in PDT. In Proceedings of the 23rd International Conference on Computational Linguistics (Coling 2010), Beijing, China, pp. 9-12. find
  9. Prasad, R., Dinesh, N., Lee, A., Miltsakaki, E., Robaldo, L., Joshi, A. and Webber, B. (2007). The Penn Discourse TreeBank 2.0 Annotation Manual. Available at: http://www.seas.upenn.edu/~pdtb/PDTBAPI/pdtb-annotation-manual.pdf. ↗
  10. Prasad, R., Dinesh, N., Lee, A., Miltsakaki, E., Robaldo, L., Joshi, A. and Webber, B. (2008). The Penn Discourse Treebank 2.0. In Proceedings of the 6th International Conference on Language Resources and Evaluation (LREC 2008), Marrakech, Morocco, pp. 2961–2968. find our card
  11. Sgall, P., Hajičová, E. and Panevová, J. (1986). The Meaning of the Sentence in Its Semantic and Pragmatic Aspects, Praha: Academia. find
  12. Zeyrek, D., Demirşahin Işın, Çallı A. B. S., Balaban H. Ö., Yalçınkaya İ., & Turan Ü. D. (2010). The annotation scheme of the Turkish Discourse Bank and an evaluation of inconsistent annotations. In Proceedings of the Fourth Linguistic Annotation Workshop, ACL 2010, Uppsala, Sweden, pp. 282–289. find
  13. Yuping Zhou and Nianwen Xue. (2012). PDTB-style discourse annotation of Chinese text. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics (ACL 2012), Jeju, Republic of Korea, pp. 69–77. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/jinova-2012-semi-automatic-annotation-intra-sentential/