Elamur ResearchβŒ•
☰
home / papers / 2011 / Modelling Discourse Relations for Arabic

Modelling Discourse Relations for Arabic

Summary

This EMNLP 2011 paper presents the first computational models of discourse relations for Arabic. It focuses on explicit discourse connectives. The authors take news texts drawn from Part 1 v2.0 of the Arabic Penn Treebank and annotate them according to Penn Discourse Treebank (PDTB) principles, modified to suit Arabic. A corpus study shows two things about Arabic news. First, adjacent sentences are usually linked by an overt connective, unlike in English. Second, these connectives are much more ambiguous than their English counterparts. The paper addresses two supervised tasks:

  • Connective recognition: deciding whether a candidate word has a discourse function.
  • Connective interpretation: assigning the relation the connective signals.

With gold syntactic features, recognition comes close to human agreement. A variant that uses only lexical, automatically derived morphological and POS features also performs reliably. This matters for languages that lack strong parsers. For interpretation, a wider feature set gives small but statistically significant gains over a strong baseline that assigns each connective its most frequent relation. The gains are concentrated on frequent, highly ambiguous connectives. Together, the work provides an initial resource and baseline methods for Arabic discourse parsing.

Key points

  • Relation inventory: a hierarchy of 17 relations, coarser than the PDTB set, with BACKGROUND and SIMILARITY added for Arabic news.
  • Connective coverage: clitics (e.g. l, b, f) and prepositions count as connectives; their second argument is often an Al-Masdar (verbal noun).
  • Class-level relation results, all 6039 connectives: most-frequent-class baseline 62.4% accuracy; connective-only model (M1) and the full model (M2) both reach 88.7%.
  • Class-level relation results, excluding sentence-initial wa (3813 tokens): baseline 41.8%, M1 82.7% (ΞΊ 0.74), M2 83.5% (ΞΊ 0.75).
  • Ambiguous connectives: 33 types account for 3239 tokens, and on these M2 beats M1 significantly at the 1% level. A single rule labelling clitic prepositions as Cause.Reason covers 494 cases with 26 errors.

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

Fields: Linguistics & languages

Topics: Discourse analysis

Cited by

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

Cite

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

@inproceedings{alsaif2011modelling,
  title = {{Modelling Discourse Relations for Arabic}},
  author = {Al-Saif, Amal and Markert, Katja},
  booktitle = {Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing},
  year = {2011},
  url = {https://aclanthology.org/D11-1068/}
}

References

38 references, 17 of them with a card of ours. β†— leads to the work, Β«findΒ» searches for one we have no id for yet.

  1. M. Abdl al latif, A. Umar, and M. Zahran. 1997. Alnhw AlAsAsi. Dar Alfker Al-Arabi, Cairo, Egypt. find
  2. A. AlSaif and K. Markert. 2010. The leeds arabic discourse treebank: Annotating discourse connectives for arabic. In Language Resources and Evaluation Conference (LREC). find our card
  3. J. Baldridge and A. Lascarides. 2005. Probabilistic head-driven parsing for discourse structure. In Proc. find
  4. S. Blair-Goldensohn, K McKeown, and O. Rambow. 2007. Building and refining rhetorical-semantic relation models. In Proc. of HLT-NAACL 2007. β†—
  5. L. Carlson, D. Marcu, and M. Okurewski. 2002. Rst discourse treebank. Linguistic Data Consortium. find
  6. D. duVerle and H. Prendinger. 2009. A novel discourse parser based on support vector machine classification. In Proc. of ACL 2009. find our card
  7. R. Elwell and J. Baldridge. 2008. Discourse connective argument identification with connective specific rankers. In Proc. of the International Conference on Semantic Computing. find
  8. R. Girju. 2003. Automatic detection of causal relations for questions answering. In Proc. of the ACL 2003 Workshop on Multilingual Summarisation and Question Answering, pages 76–83. find
  9. M.A.K. Halliday and R. Hasan. 1976. Cohesion in English. Longman London. find
  10. J.R. Hobbs. 1985. On the coherence and structure of discourse. Center for the Study of Language and Information, Stanford, Calif. find
  11. A. Knott and T. Sanders. 1998. The classification of coherence relations and their linguistic markers: An exploration of two languages. Journal of Pragmatics, 30(2):135–175. β†—
  12. Z. Lin, M. Kan, and H.T. Ng. 2009. Recognizing implicit discourse relations in the penn discourse treebank. In Proc. of EMNLP 2009, pages 343–351. find our card
  13. A. Louis and A. Nenkova. 2010. Creating local coherence: An empirical assessment. In Proc. of NAACL 2010. find
  14. M. Maamouri and A. Bies. 2004. Developing an Arabic treebank: Methods, guidelines, procedures, and tools. In Proceedings of the Workshop on Computational Approaches to Arabic Script-based Languages (COLING), Geneva. find our card
  15. W.C. Mann and S.A. Thompson. 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text, 8(3):243–281. β†— our card
  16. D. Marcu and A. Echihabi. 2002. An unsupervised approach to recognizing discourse relations. In Proc. of ACL 2002. find our card
  17. D. Marcu. 2000. The theory and practice of discourse parsing and summarization. MIT Press. β†—
  18. E. Miltsakaki, N. Dinesh, R. Prasad, A. Joshi, and B. Webber. 2005. Experiments on sense annotation and sense disambiguation of discourse connectives. In Proc. of the Workshop on Treebanks and Linguistic Theories. find
  19. E. Pitler and A. Nenkova. 2008. Revisiting readability: A unified framework for predicting text quality. In Proc. of EMNLP 2008, pages 186–195. find
  20. E. Pitler and A. Nenkova. 2009. Using syntax to disambiguate explicit discourse connectives. In Proc of ACL-IJCNLP 2009 (Short Papers), pages 13–16. find
  21. E. Pitler, M. Raghupathy, H. Mehta, A. Nenkova, A. Lee, and A. Joshi. 2008. Easily identifiable discourse relations. In Proceedings of the 22nd International Conference on Computational Linguistics (COLING 2008), Manchester, UK, August. find our card
  22. E. Pitler, A. Louis, and A. Nenkova. 2009. Automatic sense prediction for implicit discourse relations in text. In Proc. of ACL-IJCNLP 2009, pages 683–691. find our card
  23. R. Prasad, N. Dinesh, A. Lee, E. Miltsakaki, L. Robaldo, A. Joshi, and B. Webber. 2008a. The Penn discourse treebank 2.0. In Proceedings of the 6th International Conference on Language Resources and Evaluation (LREC 2008). find our card
  24. R. Prasad, S. Husain, D.M. Sharma, and A. Joshi. 2008b. Towards an Annotated Corpus of Discourse Relations in Hindi. In The Third International Joint Conference on Natural Language Processing, pages 7–12. Citeseer. find our card
  25. K.C. Ryding. 2005. A reference grammar of modern standard Arabic. Cambridge Univ Pr. β†—
  26. S. Siegel and N.J. Castellan. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill New York. find
  27. S. Somasundaran, J. Wiebe, and J. Ruppenhofer. 2008. Discourse-level opinion interpretation. In Proc. of Coling 2008. find
  28. R. Soricut and D. Marcu. 2003. Sentence level discourse parsing using syntactic and lexical information. In Proc of HLT-NAACL 2003. find our card
  29. C. Sporleder and A. Lascarides. 2008. Using automatically labelled examples to classify rhetorical relations: An assessment. Natural Language Engineering, 14:369–416. find
  30. W. Wang, J. Su, and C. Tan. 2010. Kernel-based discourse relation recognition with temporal ordering information. In Proc. of ACL 2010, pages 710–719. find our card
  31. B. Webber, A. Knott, M. Stone, and A. Joshi. 1999. Discourse relations: A structural and presuppositional account using lexicalised TAG. In Proceedings of the 37th annual meeting of the Association for Computational Linguistics on Computational Linguistics, page 48. Association for Computational Linguistics. find our card
  32. B. Wellner and J. Pustejovski. 2007. Automatically identifying the arguments of discourse connectives. In Proc. of EMNLP 2007, pages 92–101. find
  33. B. Wellner, J. Pustejovski, A. Havasi, A. Rumshisky, and R. Suair. 2006. Classification of discourse coherence relations: An exploratory study using multiple knowledge sources. In Proc. of SIGDIAL2006. find our card
  34. Janyce Wiebe, Theresa Wilson, and Claire Cardie. 2005. Annotating expressions of opinions and emotions in language. Language Resources and Evaluation. β†—
  35. W. Wright. 2008. A grammar of the Arabic language. Bibliobazaar. find
  36. Nianwen Xue. 2005. Annotating discourse connectives in the chinese treebank. In CorpusAnno ’05: Proceedings of the Workshop on Frontiers in Corpus Annotations II, pages 84–91, Morristown, NJ, USA. Association for Computational Linguistics. find our card
  37. D. Zeyrek and B. Webber. 2008. A discourse resource for turkish: Annotating discourse connectives in the metu corpus. Proceedings of IJCNLP-2008. Hyderabad, India. Z. Zhou, Y. Xu, Z. Niu, M. Lan, . Su, and Tan. find our card
  38. C. 2010. Predicting discourse connectives for implicit discourse relation recognition. In Proc. of Coling 2010, pages 1507–1514. find our card

↑ back to top

This page in Markdown β†’