Elamur Research⌕
☰
home / papers / 2012 / Measuring the Strength of Linguistic Cues for Discourse Relations

Measuring the Strength of Linguistic Cues for Discourse Relations

Summary

Discourse relations are usually labelled as either explicit, when a connective such as "because" is present, or implicit. Torabi Asr and Demberg argue that this split is too coarse, because connectives differ greatly in how clearly they signal a particular relation sense. They define a cue-strength score derived from Bayes' theorem: the ratio p(cue|r)/p(cue), which shows how much observing a marker raises the likelihood of relation r above its prior. They compute this score on the Penn Discourse Treebank (PDTB), using only relations annotated at the finest sense level and applying add-1 smoothing. Some connectives turn out to be highly specific, while very frequent ones like "and" and "but" spread their weak signal over many senses. The authors conclude that implicitness should be treated as a graded property rather than a yes/no feature. They also suggest the score could guide annotation, support fine-grained automatic sense classification, and help select weakly marked explicit relations as training data for implicit-relation classifiers.

Key points

  • The PDTB (Wall Street Journal texts) contains 18,459 explicit and 16,053 implicit relations. The study keeps only relations tagged with one of the 30 leaf senses, which yields 95 explicit and 70 implicit connective types.
  • After add-1 smoothing, strength values range from 0.0028 to 71.4370.
  • Among the 20 most frequent connectives, the strongest are "for example" for Instantiation (42.17), "although" (23.34) and "though" (18.44) for expectation, and "so" for result (20.36).
  • For reason relations, "because" scores 11.80 and "when" only 1.13. This may explain why annotators seldom inserted "when" in implicit reason relations.
  • For 45 of the 95 connectives, the relation they occur with most often differs from the relation they mark most strongly. For example, "while" appears most often in Synchrony but is a more reliable signal of opposition.

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

8 papers of our library cite it (of the 140673 whose bibliographies we read); 3 of them have a page here, 5 not yet.

Cite

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

@inproceedings{asr2012measuring,
  title = {{Measuring the Strength of Linguistic Cues for Discourse Relations}},
  author = {Asr, Fatemeh Torabi and Demberg, Vera},
  booktitle = {Proceedings of the Workshop on Advances in Discourse Analysis and its Computational Aspects},
  year = {2012},
  url = {https://aclanthology.org/W12-4703/}
}

References

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

  1. Asr, F. T. and Demberg, V. (2012). Implicitness of discourse relations. In Proceedings of COLING, Mumbai, India. find our card
  2. Carlson, L., Marcu, D., and Okurowski, M. (2003). Building a discourse-tagged corpus in the framework of rhetorical structure theory. Current and new directions in discourse and dialogue, pages 85–112. find our card
  3. Halliday, M. and Hasan, R. (1976). Cohesion in English. Longman (London). find
  4. Müller, S. (2005). Discourse markers in native and non-native english discourse. Pragmatics and Beyond, 138:1–297. find
  5. Pitler, E., Raghupathy, M., Mehta, H., Nenkova, A., Lee, A., and Joshi, A. (2008). Easily identifiable discourse relations. Technical Reports (CIS), page 884. find our card
  6. 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, pages 2961–2968. find our card
  7. Prasad, R., Joshi, A., and Webber, B. (2010). Realization of discourse relations by other means: alternative lexicalizations. In Proceedings of the 23rd International Conference on Computational Linguistics: Posters, pages 1023–1031. find our card
  8. Rohde, H. and Horton, W. (2010). Why or what next? eye movements reveal expectations about discourse direction. In Proceedings of 23rd Annual CUNY Conference on Human Sentence Processing, pages 18–20. find
  9. Sanders, T., Spooren, W., and Noordman, L. (1992). Toward a taxonomy of coherence relations. Discourse Processes, 15(1):1–35. find
  10. Sporleder, C. (2008). Lexical models to identify unmarked discourse relations: Does WordNet help? Lexical-Semantic Resources in Automated Discourse Analysis, page 20. find
  11. Stede, M. (2011). Discourse processing. Synthesis Lectures on Human Language Technologies, 4(3):1–165. find
  12. Van Dijk, T. (2004). From text grammar to critical discourse analysis. A brief academic autobiography. version, 2. find
  13. Webber, B. (2012). Alternatives and extra-propositional meaning. PASCAL 2 Invited Talk at ExProM Workshop, Jeju Island, Korea. find
  14. Zhou, Z., Xu, Y., Niu, Z., Lan, M., Su, J., and Tan, C. (2010). Predicting discourse connectives for implicit discourse relation recognition. In Proceedings of the 23rd International Conference on Computational Linguistics: Posters, pages 1507–1514. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/asr-2012-measuring-strength-linguistic-cues-discourse/