Generating Typed Dependency Parses from Phrase Structure Parses

What it is about

The paper describes a rule system that turns Penn Treebank phrase-structure trees, as produced by the Stanford parser, into typed dependencies. The method has two steps: extract head-dependent pairs using head rules adapted from Collins' rules but aimed at semantic heads (for example, a relative clause takes the verb as head), then assign one of 48 grammatical relations from a hierarchy rooted in "dependent". The hierarchy includes NP-internal relations such as nn, appos, num, number and abbrev, which the authors argue matter for applications. Optional collapsing turns prepositions and conjunctions into direct relations. On 10 Brown sentences the output scores 80.3% per-dependency accuracy; the paper also compares with Minipar and the Link parser on the same sample and calls the estimate rough, since the sample is small.

Ideas

• I1. Converter check in en/src/convert.rs: count the gold relations that come out as the generic dep where a specific one is expected (the paper's own error rule), on the 10 sample sentences and on the PTB-to-mova conversion; report per 1k tokens.
• I2. Head choice for relative clauses and flat NPs: test the paper's rule (verb as head of a relative clause, two heads for "the new phone book and tour guide") as unit tests in the converter; measure the change in relation counts on the PTB conversion (test only, PTB is NC).
• I3. Collapsed layer as a derived view: keep the collapsed preposition and conjunct relations only as a derived layer, in line with the dialect rule that Enhanced UD is derived, and count how many world events change on a 100-sentence sample.
• I4. NP-internal relations as a treeagree rule: count gate violations for nn, appos, num and abbrev on Haiku silver with and without the rule, using treeagree check.

For Mova

The relevant code is en/src/convert.rs and the converters under treebanks/en/<bank>/; the chapters are 22. Phrases: noun, adjective, adverb and prepositional phrases (NP-internal structure), 07-punctuation-capitals-numbers.md (num and number) and 23-complementation.md. Measure first: the share of dep labels in the PTB-converted corpus, and the 10-sentence sample as a fixed regression check.

Topics: Dependency parsing

Cite

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

@inproceedings{marneffe2006generating,
  title = {{Generating Typed Dependency Parses from Phrase Structure Parses}},
  author = {Marneffe, Marie-Catherine de and MacCartney, Bill and Manning, Christopher D.},
  booktitle = {Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC'06)},
  year = {2006},
  url = {https://aclanthology.org/L06-1260/}
}

References

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

1. Joan Bresnan. 2001. Lexical-Functional Syntax. Blackwell, Oxford. find
2. John Carroll, Guido Minnen, and Ted Briscoe. 1999. Corpus annotation for parser evaluation. In Proceedings of the EACL workshop on Linguistically Interpreted Corpora (LINC). find our card
3. Eugene Charniak. 2000. A maximum-entropy-inspired parser. In Proceedings of NAACL-2000. find
4. Noam Chomsky. 1981. Lectures on Government and Binding. Foris, Dordrecht. find
5. Michael Collins. 1999. Head-Driven Statistical Models for Natural Language Parsing. Ph.D. thesis, University of Pennsylvania. The Fulton County Grand Jury said Friday an investigation of Atlanta ’s recent primary election produced “no evidence” that any irregularities took place. However, the jury said it believes “these two offices should be combined to achieve greater efficiency and reduce the cost of administration”. The jury also commented on the Fulton ordinary’s court which has been under fire for its practices in the appointment of appraisers, guardians and administrators and the awarding of fees and compensation. When the larvae hatch, they feed on the beebread, although they also receive extra honey meals from their mother. In her letter to John Brown, “E. B.”, the Quakeress from Newport, had suggested that the American people owed more honor to John Brown for seeking to free the slaves than they did to George Washington. Below he could see the bright torches lighting the riverbank. “How many pamphlets do we have in stock?”, Rector said. Then Rector, attired in his best blue serge suit, sat in a chair out on the lawn, in the shade of a tree, smoking a cigarette and waiting. Have you any objection to the following plan? She was watching a tree ride wildly down that roiling current. Table 2: 10 sentences from the Brown Corpus, to compare outputs of Minipar, the Link Parser and the Stanford parser. find our card
6. Marie-Catherine de Marneffe, Bill MacCartney, Trond Grenager, Daniel Cer, Anna Rafferty, and Christopher D. Manning. 2006. Learning to distinguish valid textual entailments. To appear in P ASCAL RTE-2 Challenge workshop. find
7. Tracy H. King, Richard Crouch, Stefan Riezler, Mary Dalrymple, and Ronald Kaplan. 2003. The PARC 700 dependency bank. In 4th International Workshop on Linguistically Interpreted Corpora (LINC-03). find
8. Dan Klein and Christopher D. Manning. 2003. Accurate unlexicalized parsing. In Proceedings of the 41st Meeting of the Association for Computational Linguistics. find our card
9. Matthias T. Kromann. 2003. The Danish Dependency Treebank and the underlying linguistic theory. In Joakim Nivre and Erhard Hinrichs, editors, Proceedings of the Second Workshop on Treebanks and Linguistic Theories (TLT 2003). Vaxjo University Press. find
10. Roger Levy and Galen Andrew. 2006. Tregex and Tsurgeon: tools for querying and manipulating tree data structures. In LREC 2006. http://www-nlp.stanford.edu/software/tregex.shtml. ↗
11. Dekang Lin. 1998. Dependency-based evaluation of MINIPAR. In Workshop on the Evaluation of Parsing Systems, Granada, Spain. find
12. Philip H. Miller. 2000. Strong Generative Capacity: The Semantics of Linguistic Formalism. Number 46 in Lecture Notes. CSLI Publications, Stanford, CA. find
13. Dan Moldovan, Christine Clark, Sanda Harabagiu, and Steve Maiorano. 2003. Cogex: A logic prover for question answering. In HLT-NAACL. find
14. Rajat Raina, Andrew Y. Ng, and Christopher D. Manning. 2005. Robust textual inference via learning and abductive reasoning. In Proceedings of AAAI 2005. AAAI Press. find
15. Daniel D. Sleator and Davy Temperley. 1993. Parsing English with a link grammar. In Third International Workshop on Parsing Technologies. find our card
16. Leonoor van der Beek, Gosse Bouma, and Robert Malouf andGertjan van Noord. 2002. The Alpino Dependency Treebank. In Computational Linguistics in the Netherlands CLIN 2001. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/map-2006-marneffe-generating-typed-dependency-parses-phrase/