Text Simplification using Typed Dependencies: A Comparision of the Robustness of Different Generation Strategies

What it is about

The paper presents REGENT, a framework that applies 63 transformation rules to Stanford typed dependencies to simplify coordination, subordination, relative clauses, apposition and passive voice. It compares two ways of turning the transformed graphs back into text: gen-light reuses the word order and morphology of the original sentence and codes changes in the rules, while gen-heavy converts to DSyntS and lets the RealPro statistical realiser decide order and morphology. The test set is 175 sentences from six news articles, run on the single best parse and on 50-best parses with an overgenerate-and-rank choice. Accuracy, judged by the developer, is 83.9% for gen-light on one parse and 87.9% on 50 parses, against 70.8% and 77.7% for gen-heavy. The main failure of gen-light is relative-clause attachment error from the parser, which gives a wrong but fluent sentence; gen-heavy garbles word order.

Ideas

• I1. Keep the source order as the default in Mova's linearisation and encode only the changes. en/src/lin.rs orders words from side tables; the paper's result suggests that a default from the source, with explicit overrides counted per class, is more robust to parse errors. Measure the override rate per class in the residual.
• I2. When tools/roundtrip is unsure of an attachment, keep several trees and let the encoder pick the one whose output agrees with the other translation of the same verse. The score is a count of agreeing concepts, deterministic, run twice.
• I3. Report accuracy together with the average length of output, as the paper does, so that a change which simplifies a lot but breaks grammar cannot pass a gate.

For Mova

The contrast between gen-light and gen-heavy is the same choice as the third open problem of the translator: the encoder should write trees, not word strings. A surface realiser fed from a meta-world with no source order is brittle, as the paper shows for a statistical one. The paper's own sample is small and judged by one person, so the numbers are a warning, not a target. Summary only; no text is carried into the build.

Topics: Machine translation

Cite

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

@inproceedings{siddharthan2011text,
  title = {{Text Simplification using Typed Dependencies: A Comparision of the Robustness of Different Generation Strategies}},
  author = {Siddharthan, Advaith},
  booktitle = {Proceedings of the 13th European Workshop on Natural Language Generation},
  year = {2011},
  url = {https://aclanthology.org/W11-2802/}
}

References

33 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Richard Anderson and Peter Freebody. 1981. Vocabulary knowledge. In John Guthrie, editor, Comprehension and Teaching: Research Reviews, pages 77–117. International Reading Association, Newark, DE. find
2. Richard Anderson. 1981. A proposal to continue a center for the study of reading. Technical Report 487, University of Illinois, Center for the Study of Reading, Urbana-Champaign. find
3. Regina Barzilay and Lillian Lee. 2003. Learning to paraphrase: An unsupervised approach using multiple-sequence alignment. In HLT-NAACL 2003: Main Proceedings, pages 16–23. find our card
4. Regina Barzilay and Kathleen McKeown. 2005. Sentence fusion for multidocument news summarization. Computational Linguistics, 31(3):297–328. find
5. Isabel L. Beck, Margaret G. McKeown, Gale M. Sinatra, and Jane A. Loxterman. 1991. Revising social studies text from a text-processing perspective: Evidence of improved comprehensibility. Reading Research Quarterly, pages 251–276. find
6. John Carroll, Guido Minnen, Yvonne Canning, Siobhan Devlin, and John Tait. 1998. Practical simplification of English newspaper text to assist aphasic readers. In Proceedings of AAAI98 Workshop on Integrating Artificial Intelligence and Assistive Technology, pages 7– 10, Madison, Wisconsin. find
7. Raman Chandrasekar and Bangalore Srinivas. 1997. Automatic induction of rules for text simplification. Knowledge-Based Systems, 10:183–190. find
8. Raman Chandrasekar, Christine Doran, and Bangalore Srinivas. 1996. Motivations and methods for text simplification. In Proceedings of the 16th International Conference on Computational Linguistics (COLING ’96), pages 1041–1044, Copenhagen, Denmark. find
9. Trevor Cohn and Mirella Lapata. 2009. Sentence compression as tree transduction. Journal of Artificial Intelligence Research, 34(1):637–674. find
10. Trevor Cohn, Chris Callison-Burch, and Mirella Lapata. 2008. Constructing corpora for the development and evaluation of paraphrase systems. Computational Linguistics, 34(4):597–614. find
11. Meredyth Daneman and Patricia Carpenter. 1980. Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior, 19:450–466. find
12. Marie-Catherine De Marneffe, Bill MacCartney, and Christopher D. Manning. 2006. Generating typed dependency parses from phrase structure parses. In Proceedings of LREC, volume 6, pages 449–454. Citeseer. find our card
13. Siobhan Devlin and John Tait. 1998. The use of a psycholinguistic database in the simplification of text for aphasic readers. In J. Nerbonne, editor, Linguistic Databases, pages 161–173. CSLI Publications, Stanford, California. find
14. Michel Galley and Kathleen McKeown. 2007. Lexicalized Markov grammars for sentence compression. In Human Language Technologies 2007: The Conference of the North American Chapter of the Association for Computational Linguistics; Proceedings of the Main Conference, pages 180–187, Rochester, New York, April. Association for Computational Linguistics. find
15. Caroline Gasperin, Erick Maziero, and Sandra Aluı́sio. 2010. Challenging choices for text simplification. Computational Processing of the Portuguese Language, pages 40–50. find
16. Ali Ibrahim, Boris Katz, and Jimmy Lin. 2003. Extracting paraphrases from aligned corpora. In Proceedings of The Second International Workshop on Paraphrasing. find
17. Nobuhiro Kaji, Daisuke Kawahara, Sadao Kurohash, and Satoshi Sato. 2002. Verb paraphrase based on case frame alignment. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL’02), pages 215–222, Philadelphia, USA. find
18. Kevin Knight and Daniel Marcu. 2000. Statistics-based summarization — step one: Sentence compression. In Proceeding of The American Association for Artificial Intelligence Conference (AAAI-2000), pages 703–710. find
19. Emiel Krahmer, Erwin Marsi, and Paul van Pelt. 2008. Query-based sentence fusion is better defined and leads to more preferred results than generic sentence fusion. In Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics on Human Language Technologies: Short Papers, pages 193–196. Association for Computational Linguistics. find
20. James J. L’Allier. 1980. An evaluation study of a computer-based lesson that adjusts reading level by monitoring on task reader characteristics. Ph.D. thesis, University of Minnesota, Minneapolis, MN. find
21. Benoit Lavoie and Owen Rambow. 1997. A fast and portable realizer for text generation systems. In Proceedings of the fifth conference on Applied natural language processing, pages 265–268. Association for Computational Linguistics. find our card
22. Tracy Linderholm, Michelle G. Everson, Paul van den Broek, Maureen Mischinski, Alex Crittenden, and J. Samuels. 2000. Effects of Causal Text Revisions on More-and Less-Skilled Readers’ Comprehension of Easy and Difficult Texts. Cognition and Instruction, 18(4):525–556. find
23. Erwin Marsi and Emiel Krahmer. 2005. Explorations in sentence fusion. In Proceedings of the European Workshop on Natural Language Generation, pages 109–117. find
24. Jana Mason and Janet Kendall. 1979. Facilitating reading comprehension through text structure manipulation. Alberta Journal of Medical Psychology, 24:68– 76. find
25. George A. Miller, Richard Beckwith, Christiane D. Fellbaum, Derek Gross, and Katherine Miller. 1993. Five Papers on WordNet. Technical report, Princeton University, Princeton, N.J. find
26. Sarah E. Petersen. 2007. Natural language processing tools for reading level assessment and text simplification for bilingual education. Ph.D. thesis, University of Washington, Seattle, WA. find
27. Stephen P. Quigley and Peter V. Paul. 1984. Language and Deafness. College-Hill Press, San Diego, California. find
28. Philip Quinlan. 1992. The Oxford Psycholinguistic Database. Oxford University Press, U.K. find
29. Stefan Riezler, Tracy H. King, Richard Crouch, and Annie Zaenen. 2003. Statistical sentence condensation using ambiguity packing and stochastic disambiguation methods for lexical-functional grammar. In Proceedings of the Human Language Technology Conference and the 3rd Meeting of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL’03), Edmonton, Canada. find
30. Advaith Siddharthan and Napoleon Katsos. 2010. Reformulating discourse connectives for non-expert readers. In Proceedings of the 11th Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL-HLT 2010), Los Angeles, CA. find
31. Advaith Siddharthan. 2003. Preserving discourse structure when simplifying text. In Proceedings of the European Natural Language Generation Workshop (ENLG), 11th Conference of the European Chapter of the Association for Computational Linguistics (EACL’03), pages 103–110, Budapest, Hungary. find
32. Advaith Siddharthan. 2006. Syntactic simplification and text cohesion. Research on Language and Computation, 4(1):77–109. find
33. Advaith Siddharthan. 2010. Complex lexico-syntactic reformulation of sentences using typed dependency representations. In Proceedings of the 6th International Natural Language Generation Conference (INLG 2010), pages 125–133. Dublin, Ireland. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/siddharthan-2011-text-simplification-typed-dependencies-comparision/