PropBank: Semantics of New Predicate Types

What it is about

PropBank annotated verbs only. This paper describes how it was extended to nouns, adjectives and complex predicates such as light verb constructions (make an offer). Over 2,800 noun rolesets were added to the lexicon, many taken from NomBank and extended with WordNet classes for events and states. For a light verb construction, the light verb and the noun are annotated in two passes, and the noun's roleset is used, so make an offer is the offer roleset. For adjectives, the support verb and the adjective each get their own arguments; a "degree-consequence" construction (too or enough with an adjective and an infinitive) is marked with a construction tag, because the construction, not the adjective, projects the argument. The paper proposes aliases: fear-verb, fear-noun and afraid-adjective share one roleset, so that AMR can treat them as one event. It contrasts this with FrameNet, which places fear-noun and fear-verb in different frames. The paper reports practice and design, not measured results.

Ideas

• I1. Aliases as a table: nouns and adjectives that share a sense with a verb map to one roleset ID (the fear.01 pattern) in en/src/dict.rs. Measure how many school-lesson nouns and adjectives gain a roleset through an alias.
• I2. Light-verb rule: when a light verb (make, take, have, give, do) takes a noun object that has a roleset, the event is the noun's roleset, and the light verb's subject is the noun's first argument. Test on "make an offer" and similar school sentences in prag/src/roles.rs.
• I3. Degree-consequence as a closed pattern: "too or enough" plus an adjective plus an infinitive is marked so that the infinitive is an argument of the construction and not of the adjective. The rule goes in prag/src/roles.rs, and the test is school sentences with too ... to and enough ... to.

For Mova

The chapters are 22. Phrases: noun, adjective, adverb and prepositional phrases, 15. Adjectives and adverbs and 19. The lexicon: vocabulary, multiword expressions and register. The code is en/src/dict.rs and prag/src/roles.rs. The measure is the proportion of school sentences with a light verb or an adjective construction whose roles come out right, run twice to confirm identical output.

Topics: Semantic role labeling

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.63317/444ookt3fyf5

@inproceedings{bonial2014propbank,
  title = {{PropBank: Semantics of New Predicate Types}},
  author = {Bonial, Claire and Bonn, Julia and Conger, Kathryn and Hwang, Jena D. and Palmer, Martha},
  booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)},
  year = {2014},
  doi = {10.63317/444ookt3fyf5},
  url = {https://doi.org/10.63317/444ookt3fyf5}
}

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. Laura Banarescu, Claire Bonial, Shu Cai, Madalina Georgescu, Kira Griffitt, Ulf Herjakob, Kevin Knight, Philipp Kohen, Martha Palmer, and Nathan Schneider. 2013. Abstract meaning representation for sembanking. In Proceedings of the 7th Linguistic Annotation Workshop and Interoperability with Discourse. find our card
2. Cristiane Fellbaum, Joachim Grabowski, and Shari Landes. 1998. Performance and confidence in a semantic annotation task. In Christiane Fellbaum, editor, WordNet: An Electronic Database. The MIT Press. find
3. Charles J. Fillmore, Russell R. Lee-Goldman, and Russell Rhomieux. 2012. The FrameNet Constructicon. In Hans C. Boas and Ivan A. Sag, editors, Sign-Based Construction Grammar. CSLI Publications, October. find
4. Ana-Maria Giuglea and Alessandro Moschitti. 2006. Semantic role labelling via FrameNet, VerbNet and Prop-Bank. In Proceeding of COLING/ACL-2006, Sydney, Australia. find
5. Adele E. Goldberg. 1995. Constructions: A Construction Grammar Approach to Argument Structure. University Of Chicago Press. find
6. Jena D. Hwang, Archna Bhatia, Claire Bonial, Aous Mansouri, Ashwini Vaidya, Nianwen Xue, and Martha Palmer. 2010. Propbank annotation of multilingual light verb constructions. In Proceedings of the Fourth Linguistic Annotation Workshop, pages 82–90, Uppsala, Sweden, July. Association for Computational Linguistics. find
7. Edward Loper, Szu-ting Yi, and Martha Palmer. 2007. Combining lexical resources: Mapping between Prop-Bank and VerbNet. In Proceedings of the Seventh International Workshop on Computational Semantics (IWCS-7). find
8. Mitchell P. Marcus, Beatrice Santorini, and Mary Ann Marcinkiewicz. 1994. Building a large annotated corpus of english: the penn treebank. Computational Linguistics, 19. find our card
9. Paola Merlo and Lonneke van der Plas. 2009. Abstraction and generalisation in semantic role labels: Propbank, verbnet or both? In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP, pages 288–296, Suntec, Singapore, August. find our card
10. 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 the NAACL/HLT Workshop on Frontiers in Corpus Annotation. find
11. Gerhard Nickel. 1978. Complex verbal structures in english. In Dietrich Nehls, editor, Studies in Descriptive English Grammar, pages 63–83. Julius Groos. find
12. Martha Palmer, Dan Guildea, and Paul Kingsbury. 2005. The Proposition Bank: An annotated corpus of semantic roles. Computational Linguistics, 31(1):71–105, March. find our card
13. Martha Palmer. 2009. Semlink: Linking PropBank, Verb-Net and FrameNet. In Proceedings of the Generative Lexicon Conference (GenLex-09), Pisa, Italy. find
14. Lei Shi and Rada Mihalcea. 2005. Putting pieces together: Combining framenet, verbnet and wordnet for robust semantic parsing. In Computational Linguistics and Intelligent Text Processing; Sixth International Conference (CICLing 2005), pages 100–111. find
15. Suzanne Stevenson, Afsaneh Fazly, and Ryan North. 2004. Statistical measures of the semi-productivity of light verb constructions. In Proceedings of the ACL 2004 Workshop on Multiword Expressions: Integrating Processing. find
16. Szu-ting Yi, Edward Loper, and Martha Palmer. 2007. Can semantic roles generalize across genre. In Proceedings of HLT/NAACL-2007. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/bonial-2014-propbank-semantics-new-predicate-types/