Prepositional Phrase Attachment Problem Revisited: How VERBNET Can Help

What it is about

The authors add lexical-semantic knowledge from VerbNet and WordNet to PP attachment. For each P-phrase (verb, noun, preposition, noun) they build features from the verb's VerbNet class and from selectional constructs, which state whether a thematic role such as instrument fits the verb and the noun. The classifier is logistic regression, scored by 10-fold cross-validation on the 23,898 tuples of the Ratnaparkhi, Reynar and Roukos (1994) data. Their own per-preposition baseline reaches 74.6% on all tuples, and the full system reaches 79.3%. For the preposition "with", dropping low-precision features raises the score from 70.7% to 72.0%. The authors report that the feature "VerbNet[noun1, noun2]" has at least 83.6% precision for six of ten prepositions.

Ideas

• I1. world/src/frames.rs: turn each VerbNet thematic role and selectional construct that fits a verb-noun pair into an explicit boolean feature, and print which construct fired. Measure accuracy per preposition against a per-preposition baseline on UD EWT PP attachments.
• I2. Keep one count table per preposition, since the paper's prepositions are strongly biased (99% of "of" goes to the noun in this data). Measure whether the per-preposition table beats a single global table on the EWT test subset.
• I3. Report each feature's precision and recall, as the paper's Figure 4 does, and drop features below a fixed precision threshold chosen on dev. Run twice to confirm the same features survive.

For Mova

The chapters are 22. Phrases: noun, adjective, adverb and prepositional phrases and 23-complementation.md. The code is world/src/frames.rs, onto/src/wsd.rs and en/src/expert.rs. The Ratnaparkhi data is derived from the Penn Treebank: test only. VerbNet's licence must be checked before any of its classes enter what is built; WordNet is open. Measure on UD EWT first.

Topics: Semantic role labeling

Cite

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

@inproceedings{bailey2015prepositional,
  title = {{Prepositional Phrase Attachment Problem Revisited: How VERBNET Can Help}},
  author = {Bailey, Daniel and Lierler, Yuliya and Susman, Benjamin},
  booktitle = {Proceedings of the 11th International Conference on Computational Semantics},
  year = {2015},
  url = {https://aclanthology.org/W15-0102/}
}

References

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

1. Agirre, E., T. Baldwin, and D. Martı́nez (2008). Improving parsing and pp attachment performance with sense information. In 46th Annual Meeting of the Association for Computational Linguistics (ACL), pp. 317–325. find
2. Agirre, E., K. Bengoetxea, K. Gojenola, and J. Nivre (2011). Improving dependency parsing with semantic classes. In 49th Annual Meeting of the Association for Computational Linguistics (ACL, Short Papers), pp. 699–703. find
3. Allen, J. (1995). Natural Language Understanding (2Nd Ed.). Redwood City, CA, USA: Benjamin-Cummings Publishing Co., Inc. find
4. Belinkov, Y., T. Lei, R. Barzilay, and A. Globerson (2014). Exploring Compositional Architectures and Word Vector Representations for Prepositional Phrase Attachment. Transactions of the Association for Computational Linguistics 2, 561–572. find our card
5. Brill, E. and P. Resnik (1994). A rule-based approach to prepositional phrase attachment disambiguation. In 15th conference on Computational linguistics-Volume 2, pp. 1198–1204. find
6. Budanitsky, A. and G. Hirst (2006, March). Evaluating wordnet-based measures of lexical semantic relatedness. Computional Linguistics 32(1), 13–47. find
7. Collins, M. and J. Brooks (1995). Prepositional phrase attachment through a backed-off model. In Proceedings of the Third Workshop on Very Large Corpora, pp. 27–38. find our card
8. Dahlgren, K. (1988). Naive Semantics for Natural Language Understanding. Norwell, MA, USA: Kluwer Academic Publishers. find
9. Ford, Bresnan, K. (1982). A competence-based theory of syntatic closure. In Bresnan (Ed.), The Mental Representation of Grammatical Relations, pp. 727–796. The MIT Press. find
10. Hall, M., E. Frank, G. Holmes, B. Pfahringer, P. Reutemann, and I. H. Witten (2009, November). The weka data mining software: An update. SIGKDD Explor. Newsl. 11(1), 10–18. find
11. Hirst, G. (1988, March). Semantic interpretation and ambiguity. Artificial intelligence 34(2), 131–177. find
12. Jensen, K. and J.-L. Binot (1987). Disambiguating prepositional phrase attachments by using on-line dictionary definitions. Computational Linguistics 13(3-4), 251–260. find
13. Katz, J. J. and J. A. Fodor (1963). The structure of a semantic theory. Language 39(2), pp. 170–210. find
14. Kipper, K., H. T. Dang, and M. Palmer (2000). Class-based construction of a verb lexicon. In 7th National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence, pp. 691–696. AAAI Press / The MIT Press. find
15. Kipper-Schuler, K. (2005). VerbNet: A Broad-Coverage, Comprehensive Verb Lexicon. Ph. D. thesis, University of Pennsylvania. find
16. Lapata, M. and F. Keller (2005). Web-based models for natural language processing. ACM Transactions on Speech and Language Processing (TSLP) 2(1), 3. find
17. Leacock, C. and M. Chodorow (1998). Combining local context and wordnet similarity for word sense identification. In C. Fellbaum (Ed.), WordNet: An Electronic Lexical Database, Cambridge, MA, USA, pp. 265–283. The MIT Press. find
18. Levin, B. (1993). English verb classes and alternations : a preliminary investigation. University Of Chicago Press. find
19. Lierler, Y. and P. Schüller (2013). Towards a tight integration of syntactic parsing with semantic disambiguation by means of declarative programming. In 10th International Conference on Computational Semantics (IWCS). find
20. Macleod, C., R. Grishman, A. Meyers, L. Barrett, and R. Reeves (1998). Nomlex: A lexicon of nominalizations. Proceedings og EURALEX 98, 187–193. find
21. Miller, G. A., R. Beckwith, C. Fellbaum, D. Gross, and K. Miller (1990). Wordnet: An on-line lexical database. International Journal of Lexicography 3, 235–244. find
22. Pantel, P. and D. Lin (2000). An unsupervised approach to prepositional phrase attachment using contextually similar words. In 38th Annual Meeting on Association for Computational Linguistics (ACL), Stroudsburg, PA, USA, pp. 101–108. find our card
23. Ratnaparkhi, A. (1998). Statistical models for unsupervised prepositional phrase attachment. In 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics - Volume 2 (ACL), Stroudsburg, PA, USA, pp. 1079–1085. find
24. Ratnaparkhi, A., J. Reynar, and S. Roukos (1994). A maximum entropy model for prepositional phrase attachment. In Workshop on Human Language Technology, pp. 250–255. find our card
25. Srikumar, V. and D. Roth (2013). Modeling semantic relations expressed by prepositions. TACL 1, 231–242. find
26. Stetina, J. and M. Nagao (1997). Corpus based pp attachment ambiguity resolution with a semantic dictionary. In 5th Workshop pn Very Large Corpora, pp. 66–80. find
27. Zapirain, B., E. Agirre, L. Màrquez, and M. Surdeanu (2013). Selectional preferences for semantic role classification. Computational Linguistics 39(3), 631–663. find
28. Zavrel, J., W. Daelemans, J. Veenstra, et al. (1997). Resolving PP attachment ambiguities with memory-based learning. In Workshop on Computational Language Learning (CoNLL’97), ACL, Madrid. find
29. Zhou, G., J. Zhao, K. Liu, and L. Cai (2011). Exploiting web-derived selectional preference to improve statistical dependency parsing. In 49th Annual Meeting of the Association for Computational Linguistics (ACL), pp. 1556–1565. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/lingbulk-2015-bailey-prepositional-phrase-attachment-problem-revisited/