Elamur Research⌕
☰
home / papers / 2004 / Inferring Sentence-internal Temporal Relations

Inferring Sentence-internal Temporal Relations

Summary

Much temporal information in text is implicit. A reader can tell which of two events came first even without an explicit cue. Lapata and Lascarides ask whether such ordering knowledge can be learned without hand-annotated temporal data. They use sentences in which a main clause and a subordinate clause are linked by an overt temporal conjunction, harvested from the machine-parsed BLLIP corpus. The marker then serves as a free label. A naive-Bayes-style probabilistic model learns, from features of both clauses, which marker connects them. Features include a "temporal signature" covering finiteness, modality, aspect, voice and polarity, along with lexical and syntactic cues.

The model is tested on two tasks:

  • Interpretation: recovering the deleted marker.
  • Sentence fusion: deciding which clause the marker attaches to, and so how the clauses are ordered.

The best configuration reaches 70.7% on interpretation and 97.4% on fusion. An ensemble of models agrees with the gold standard about as well as human judges do. The work shows that unannotated parsed corpora can support temporal inference and text generation for summarisation and question answering.

Key points

  • The data comes from BLLIP, a 30M-word Treebank-style parse of WSJ text (1987–89) produced with Charniak's (2000) parser. Clause pairs were extracted via SBAR-TMP nodes and their dominating S nodes.
  • Eight temporal conjunctions make up the confusion set: since, until, once, as, when, while, before and after. Markers rarer than 10 per million were excluded. Clause features are assumed conditionally independent given the marker, and add-k smoothing handles zero counts.
  • Human studies were run online: 198 native English speakers did the interpretation task and 100 did the fusion task.
  • On interpretation, ensemble–gold Kappa was .413, compared with .421 for humans against gold. On fusion, the ensemble scored .489 against .522 for humans.
  • Humans struggled most with as and once clauses, matching gold for only 33% and 35% of items respectively. Both lexical and syntactic features proved useful.

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

Cited by

21 papers of our library cite it (of the 140673 whose bibliographies we read); 5 of them have a page here, 16 not yet.

Cite

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

@inproceedings{lapata2004inferring,
  title = {{Inferring Sentence-internal Temporal Relations}},
  author = {Lapata, Mirella and Lascarides, Alex},
  booktitle = {Proceedings of the Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics: HLT-NAACL 2004},
  year = {2004},
  url = {https://aclanthology.org/N04-1020/}
}

References

23 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Asher, Nicholas and Alex Lascarides. 2003. Logics of Conversation. Cambridge University Press. find
  2. Barzilay, Regina. 2003. Information Fusion for Multi-Document Summarization: Praphrasing and Generation. Ph.D. thesis, Columbia University. find
  3. Charniak, Eugene. 2000. A maximum-entropy-inspired parser. In Proceedings of the 1st Conference of the North American Chapter of the Association for Computational Linguistics. Seattle, WA, pages 132–139. find our card
  4. Dietterich, T. G. 1997. Machine learning research: Four current directions. AI Magazine 18(4):97–136. find
  5. Dorr, Bonnie and Terry Gaasterland. 1995. Selecting tense aspect and connecting words in language generation. In Proceedings of 14th International Joint Conference on Artificial Intelligence. Montréal, Canada, pages 1299–1307. find
  6. Fellbaum, Christiane, editor. 1998. WordNet: An Electronic Database. MIT Press, Cambridge, MA. find
  7. Ferro, Lisa, Inderjeet Mani, Beth Sundheim, and George Wilson. 2000. Tides temporal annotation guidelines. Technical report, The MITRE Corporation. find
  8. Johnson, W. E. 1932. Probability: The deductive and inductive problems. Mind 49:409–423. find
  9. Kamp, Hans and Uwe Reyle. 1993. From Discourse to Logic: Introduction to Modeltheoretic Semantics of Natural Language, Formal Logic and Discourse Representation Theory. Kluwer, Dordrecht. find
  10. Katz, Graham and Fabrizio Arosio. 2001. The annotation of temporal information in natural language sentences. In Proceedings of ACL Workshop on Temporal and Spatial Information Processing. Toulouse, France, pages 104–111. find
  11. Lapata, Maria and Chris Brew. 1999. Using subcategorization to resolve verb class ambiguity. In Proceedings of the Joint SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora. College Park, MD, pages 266–274. find
  12. Levin, Beth. 1993. English Verb Classes and Alternations: A Preliminary Investigation. University of Chicago Press, Chicago. find
  13. Mani, Inderjeet, Barry Schiffman, and Jianping Zhang. 2003. Inferring temporal ordering of events in news. In Proceedings of the 1st Human Language Technology Conference and Annual Meeting of the North American Chapter of the Association for Computational Linguistics. Edmonton, Canada. find
  14. Marcu, Daniel and Abdessamad Echihabi. 2002. An unsupervised approach to recognizing discourse relations. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics. Philadelphia, PA, pages 368–375. find our card
  15. Quinlan, Ross J. 1993. C4.5: Programs for Machine Learning. Series in Machine Learning. Morgan Kaufman, San Mateo, CA. find our card
  16. Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. 1985. A Comprehensive Grammar of the English Language. Longman, London. find
  17. Schilder, Frank and Christopher Habel. 2001. From temporal expressions to temporal information: Semantic tagging of news messages. In Proceedings of ACL Workshop on Temporal and Spatial Information Processing. Toulouse, pages 65–72. find
  18. Setzer, Andrea and Robert Gaizauskas. 2001. A pilot study on annotating temporal relations in text. In Proceedings of ACL Workshop on Temporal and Spatial Information Processing. Toulouse, France, pages 73–80. find
  19. Siegel, Sidney and N. John Castellan. 1988. Non Parametric Statistics for the Behavioral Sciences. McGraw-Hill, New York. find
  20. Soricut, Radu and Daniel Marcu. 2003. Sentence level discourse parsing using syntactic and lexical information. In Proceedings of the 1st Human Language Technology Conference and Annual Meeting of the North American Chapter of the Association for Computational Linguistics. Edmonton, Canada, pages 22… find our card
  21. Webber, Bonnie Lynn. 1991. Structure and ostension in the interpretation of discourse deixis. Language and Cognitive Processes 6(2):107–135. find
  22. Wiebe, Janyce M., Thomas P. O’Hara, Thorsten Öhrström Sand-gren, and Kenneth J. McKeever. 1998. An empirical approach to temporal reference resolution. Journal of Artifical Intelligence Research 9:247–293. find
  23. Wilson, George, Interjeet Mani, Beth Sundheim, and Lisa Ferro. 2001. A multilingual approach to annotating and extracting temporal information. In Proceedings of ACL Workshop on Temporal and Spatial Information Processing. Toulouse, France, pages 81–87. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/lapata-2004-inferring-sentence-internal-temporal-relations/