Semantic Dependency Parsing of NomBank and PropBank: An Efficient Integrated Approach via a Large-scale Feature Selection

What it is about

The paper builds one integrated semantic dependency parser for both verbal (PropBank) and nominal (NomBank) predicates, using word-pair classification with a maximum entropy model. Two devices make it efficient: adaptive argument pruning, which adds assistant labels such as "NoMoreArgument" so that the scan over the tree stops early, and a greedy feature selection over 781 feature templates whose cost is linear in the number of templates. The pruning removes about half of the argument candidates on the syntactic path (2.13M to 1.05M training samples, 49.30%) and keeps 98.4% of gold arguments with gold syntax. On the CoNLL-2008 test set the semantic labeled F1 is 80.94 with the Johansson parser and 80.53 with the second parser, and 84.57 with gold syntax; the syntactic path beats the linear path in every setting.

Ideas

• I1. Adaptive pruning on UD trees: from the predicate, walk up to the root and collect the children left to right, stopping when a boundary label fires. Implement it in prag/src/roles.rs, and report the candidate count and the share of gold arguments kept, on the school lessons, run twice.
• I2. Greedy feature selection for the perceptron features: keep the template list in a BTreeMap, add or remove one template per round on dev, and stop when no change helps. Put it in the feature sets of en/src/ptag.rs or en/src/parse.rs. Measure dev, never test, and record the training cost.
• I3. Report the semantic F1 divided by the parser LAS, as the paper does, to see how far role quality follows parse quality. Use en parse-eval for the LAS.
• I4. Nominal predicates are the paper's second half. Use the noun chapter 11. Nouns to list nominal predicates, and test whether the same candidate pruning works for them.

For Mova

The chapters are 13. Verbs, 11. Nouns and 22. Phrases: noun, adjective, adverb and prepositional phrases. The code is prag/src/roles.rs for pruning and en/src/parse.rs for the feature sets. The measures are the candidate count, the coverage of gold arguments, and dev F1, all run twice for determinism.

Topics: Dependency parsing · Semantic role labeling

Cite

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

@inproceedings{zhao2009semantic,
  title = {{Semantic Dependency Parsing of NomBank and PropBank: An Efficient Integrated Approach via a Large-scale Feature Selection}},
  author = {Zhao, Hai and Chen, Wenliang and Kit, Chunyu},
  booktitle = {Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing},
  year = {2009},
  url = {https://aclanthology.org/D09-1004/}
}

References

30 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Xavier Carreras and Lluis Marquez. 2005. Introduction to the conll-2005 shared task: Semantic role labeling. In Proceedings of CoNLL-2005, pages 152– 164, Ann Arbor, Michigan, USA. find our card
2. Wanxiang Che, Zhenghua Li, Yuxuan Hu, Yongqiang Li, Bing Qin, Ting Liu, and Sheng Li. 2008. A cascaded syntactic and semantic dependency parsing system. In Proceedings of CoNLL-2008, pages 238–242, Manchester, England, August. find
3. Wenliang Chen, Daisuke Kawahara, Kiyotaka Uchimoto, Yujie Zhang, and Hitoshi Isahara. 2008. Dependency parsing with short dependency relations in unlabeled data. In Proceedings of IJCNLP-2008, Hyderabad, India, January 8-10. find
4. Massimiliano Ciaramita, Giuseppe Attardi, Felice Dell’Orletta, and Mihai Surdeanu. 2008. Desrl: A linear-time semantic role labeling system. In Proceedings of CoNLL-2008, pages 258–262, Manchester, England, August. find
5. Hoa Trang Dang and Martha Palmer. 2005. The role of semantic roles in disambiguating verb senses. In Proceedings of ACL-2005, pages 42–49, Ann Arbor, USA. find
6. Weiwei Ding and Baobao Chang. 2008. Improving chinese semantic role classification with hierarchical feature selection strategy. In Proceedings of EMNLP-2008, pages 324–323, Honolulu, USA. find
7. Daniel Gildea and Daniel Jurafsky. 2002. Automatic labeling of semantic roles. Computational Linguistics, 28(3):245–288. Jan Hajič, Massimiliano Ciaramita, Richard Johansson, Daisuke Kawahara, Maria Antònia Martı́, Lluı́s pred-F 1 b argu-F 1 c Verb-F 1 d Nomi-F 1 e Màrquez, Adam Meyers, Joakim Nivre, Sebastian Padó, Jan Štěpánek, Pavel Straňák, Mihai Surdeanu, Nianwen Xue, and Yi Zhang. 2009. The CoNLL-2009 shared task: Syntactic and semantic dependencies in multiple languages. In Proceedings of the 13th Conference on Computational Natural Language Learning (CoNLL-2009), June 4-5, pages 1– 18, Boulder, Colorado, USA. find our card
8. James Henderson, Paola Merlo, Gabriele Musillo, and Ivan Titov. 2008. A latent variable model of synchronous parsing for syntactic and semantic dependencies. In Proceedings of CoNLL-2008, pages 178–182, Manchester, England, August. find
9. Zheng Ping Jiang and Hwee Tou Ng. 2006. Semantic role labeling of nombank: A maximum entropy approach. In Proceedings of EMNLP-2006, pages 138–145, Sydney, Australia. find
10. Richard Johansson and Pierre Nugues. 2008. Dependency-based syntactic–semantic analysis with propbank and nombank. In Proceedings of CoNLL-2008, page 183–187, Manchester, UK. find
11. Terry Koo, Xavier Carreras, and Michael Collins. 2008. Simple semi-supervised dependency parsing. In Proceedings of ACL-08: HLT, pages 595–603, Columbus, Ohio, USA, June. find
12. Peter Koomen, Vasin Punyakanok, Dan Roth, and Wen tau Yih. 2005. Generalized inference with multiple semantic role labeling systems. In Proceedings of CoNLL-2005, pages 181–184, Ann Arbor, Michigan, USA. find
13. Chang Liu and Hwee Tou Ng. 2007. Learning predictive structures for semantic role labeling of nombank. In Proceedings of ACL-2007, pages 208–215, Prague, Czech. find
14. Lluis Marquez, Mihai Surdeanu, Pere Comas, and Jordi Turmo. 2005. A robust combination strategy for semantic role labeling. In Proceedings of HLT/EMNLP-2005, page 644–651, Vancouver, Canada. find
15. Ryan McDonald, Kevin Lerman, and Fernando Pereira. 2006. Multilingual dependency analysis with a two-stage discriminative parser. In Proceedings of CoNLL-X, New York City, June. find
16. Adam Meyers, Ruth Reeves, Catherine Macleod, Rachel Szekely, Veronika Zielinska, Brian Young, and Ralph Grishman. 2004. The nombank project: An interim report. In Proceedings of HLT/NAACL Workshop on Frontiers in Corpus Annotation, pages 24–31, Boston, Massachusetts, USA, May 6. find
17. Joakim Nivre and Ryan McDonald. 2008. Integrating graph-based and transition-based dependency parsers. In Proceedings of ACL-08: HLT, pages 950–958, Columbus, Ohio, June. find
18. Martha Palmer, Daniel Gildea, and Paul Kingsbury. 2005. The proposition bank: An annotated corpus of semantic roles. Computational Linguistics, 31(1):71–106. ↗ our card
19. Sameer Pradhan, Wayne Ward, Kadri Hacioglu, James H. Martin, and Dan Jurafsky. 2004. Shallow semantic parsing using support vector machines. In Proceedings of HLT/NAACL-2004, pages 233–240, Boston, Massachusetts, USA. find
20. Sameer Pradhan, Wayne Ward, Kadri Hacioglu, James H. Martin, and Daniel Jurafsky. 2005. Semantic role labeling using different syntactic views. In Proceedings of ACL-2005, pages 581–588, Ann Arbor, USA. find
21. James Pustejovsky, Adam Meyers, Martha Palmer, and Massimo Poesio. 2005. Merging propbank, nombank, timebank, penn discourse treebank and coreference. In Proceedings of the Workshop on Frontiers in Corpus Annotations II: Pie in the Sky, pages 5–12, Ann Arbor, USA. find our card
22. Mihai Surdeanu, Lluis Marquez, Xavier Carreras, and Pere R. Comas. 2007. Combination strategies for semantic role labeling. Journal of Artificial Intelligence Research, 29:105–151. find
23. Mihai Surdeanu, Richard Johansson, Adam Meyers, Lluı́s Màrquez, and Joakim Nivre. 2008. The CoNLL-2008 shared task on joint parsing of syntactic and semantic dependencies. In Proceedings of CoNLL-2008, pages 159–177, Manchester, UK. find our card
24. Ivan Titov, James Henderson, Paola Merlo, and Gabriele Musillo. 2009. Online graph planarisation for synchronous parsing of semantic and syntactic dependencies. In IJCAI-2009, Pasadena, California, USA. find
25. Kristina Toutanova, Aria Haghighi, and Christopher D. Manning. 2005. Joint learning improves semantic role labeling. In Proceedings of ACL-2005, pages 589–596, Ann Arbor, USA. find
26. Nianwen Xue and Martha Palmer. 2004. Calibrating features for semantic role labeling. In Proceedings of EMNLP-2004, pages 88–94, Barcelona, Spain, July 25-26. find
27. Nianwen Xue. 2006. Semantic role labeling of nominalized predicates in chinese. In Proceedings of NAACL-2006, pages 431–438, New York City, USA, June. find
28. Hai Zhao and Chunyu Kit. 2008. Parsing syntactic and semantic dependencies with two single-stage maximum entropy models. In Proceeding of CoNLL-2008, pages 203–207, Manchester, UK. find
29. Hai Zhao, Wenliang Chen, Jun’ichi Kazama, Kiyotaka Uchimoto, and Kentaro Torisawa. 2009a. Multilingual dependency learning: Exploiting rich features for tagging syntactic and semantic dependencies. In Proceedings of the 13th Conference on Computational Natural Language Learning (CoNLL-2009), June 4-5, pages 61–66, Boulder, Colorado, USA. find
30. Hai Zhao, Wenliang Chen, Chunyu Kit, and Guodong Zhou. 2009b. Multilingual dependency learning: A huge feature engineering method to semantic dependency parsing. In Proceedings of CoNLL-2009, pages 55–60, Boulder, Colorado, USA. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/zhao-2009-semantic-dependency-parsing-nombank-propbank/