Semantic Tagging for the Penn Treebank

What it is about

The paper describes a pilot in which running Penn Treebank text is tagged with WordNet senses, following the method of the SENSEVAL exercise. Two annotators tag the verbs and the head words of their noun arguments. A first annotator makes one pass, a second checks the result, and disagreements are marked as unresolved. In the pilot about 2,100 words were sense-tagged, with an overall agreement of 89%; for the 700 verbs, agreement was 81%, and 90 of the 350 verb lemmas had a disagreement. The authors revise WordNet senses when the annotators cannot separate them, and they describe criteria for sense distinctions with the example of shake, which WordNet and the Hector resource split differently. The Treebank brackets are also used to add predicate-argument structure automatically with tree patterns; on 65 sentences, 132 of 162 structures (81%) were correct, and hand correction of the 5,000-word corpus took one day.

Ideas

• I1. prag/src/roles.rs: write the predicate-argument tree patterns over UD trees (subject, object, indirect object) as explicit rules with the sense of the verb as a condition. Measure precision on a sample of sentences, the way the paper does with 65 sentences.
• I2. world/src/dictionary.rs: keep the unresolved-sense marker, as the paper's ? tag, for senses the rules cannot separate. Exclude those tokens from the accuracy count and report them as their own number.
• I3. Two passes by two authors on the same lesson sentences, with agreement reported before any discussion; the lemmas with disagreements become the list of senses to revise.

For Mova

The chapters are 19. The lexicon: vocabulary, multiword expressions and register (senses and multi-word items) and 28. Voice and the passive (argument roles). The code is prag/src/roles.rs and world/src/dictionary.rs. The measures: predicate-argument precision on a sample, the share of unresolved senses, and the two-pass agreement. The paper's data are NC-licensed and are used here only as a summary.

Topics: Treebanks

Cite

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

@inproceedings{palmer2000semantic,
  title = {{Semantic Tagging for the Penn Treebank}},
  author = {Palmer, Martha and Dang, Hoa Trang and Rosenzweig, Joseph},
  booktitle = {Proceedings of the Second International Conference on Language Resources and Evaluation (LREC'00)},
  year = {2000},
  url = {https://aclanthology.org/L00-1148/}
}

References

16 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Atkins, Sue, 1993. Tools for computer-aided corpus lexicography: The hector project. Acta Linguistica Hungarica, 41:5–72. find
2. Collins, M., 1997. Three generative, lexicalised models for statistical parsing. In Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics. Madrid, Spain. find
3. Copestake, Ann and Antonio Sanfilippo, 1993. Multilingual lexical representation. In Proceedings of the AAAI Spring Symposium: Building Lexicons for Machine Translation. Stanford, California. find
4. Dang, Hoa Trang, Karin Kipper, Martha Palmer, and Joseph Rosenzweig, 1998. Investigating regular sense extensions based on intersective levin classes. In Proceedings of Coling-ACL98. Montreal, CA. find
5. Dorr, Bonnie J., 1997. Large-scale dictionary construction for foreign language tutoring and interlingual machine translation. Machine Translation, 12:1–55. find
6. Hovy, E., 1993. Ontologies for mt. Presented at the Japan - US MT Workshop, Washington, D.C. find
7. Kilgarriff, Adam and Martha Palmer, 2000. Senseval: Evaluating word sense disambiguation systems. Computers and the Humanities, 34(1-2). find
8. Levin, Beth, 1993. English Verb Classes and Alternations A Preliminary Investigation. find
9. Lowe, J.B., C.F. Baker, and C.J. Fillmore, 1997. A frame-semantic approach to semantic annotation. In Proceedings 1997 Siglex Workshop/ANLP97. Washington, D.C. find
10. Marcus, Mitch, 1994. The penn treebank: A revised corpus design for extracting predicate argument structure. In Proceedings of the ARPA Human Language Technology Workshop. Princeton, NJ. find
11. Mel’cuk, I. A, 1988. Semantic description of lexical units in an explanatory combinatorial dictionary: Basic principles and heuristic criteria. International Journal of Lexicography, I:3:165–188. find
12. Miller, G., R. Beckwith, C. Fellbaum, D. Gross, and K. Miller, 1990. Five papers on wordnet. Technical Report 43, Cognitive Science Laboratory, Princeton University. find
13. Nirenburg, S., J. Carbonell, M. Tomita, and K. Goodman, 1992. Machine translation: a knowledge-based approach. San Mateo, California, USA: Morgan Kaufmann. find
14. Palmer, Martha, 1999. Consistent criteria for sense distinctions. Computers and the Humanities. Special Issue on SENSEVAL. find
15. Pustejovsky, James, 1991. The generative lexicon. Computational Linguistics, 17(4). find
16. XTAG-Group, The, 1995. A Lexicalized Tree Adjoining Grammar for English. Technical Report IRCS 95-03, University of Pennsylvania. find our card

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/palmer-2000-semantic-tagging-penn-treebank/