Elamur Research⌕
☰
home / papers / 2008 / Combining Word Sense and Usage for Modeling Frame Semantics

Combining Word Sense and Usage for Modeling Frame Semantics

Summary

FrameNet's limited lexical coverage and weak multilingual support constrain its use in tasks such as question answering and textual entailment. The authors propose a computational model of FrameNet that represents frames and lexical units (LUs) so that similarity between them can be measured. The model combines two sources of evidence. The paradigmatic component maps each frame onto sub-hierarchies of WordNet, using conceptual density for nouns, co-hyponymy for verbs and synonymy for adjectives. The distributional component places frames and LUs as co-occurrence vectors in a corpus-derived semantic space. The combined model supports three tasks: assigning unknown LUs to frames, reducing LU polysemy by linking LUs to WordNet synsets, and transferring English LUs to Italian via MultiWordNet. Merging the two kinds of evidence outperforms either one used alone for LU induction in both English and Italian. The Italian experiment yielded about 15,000 automatically generated LUs, offering a practical basis for building an Italian FrameNet and for assisting lexicographers in other languages.

Key points

  • Paradigmatic model. Each frame receives a separate "WordNet model" for each part of speech, built on WordNet 2.0. For nouns, a greedy conceptual-density algorithm selects the synsets that best generalize the frame's LUs.
  • Three tasks. The framework is applied to:
    • inducing LUs that FrameNet does not yet contain,
    • disambiguating LUs against WordNet senses,
    • porting LUs from English to Italian.
  • Italian manual validation, frequent words. On 1,500 word–frame decisions covering 776 words, accuracy was 0.79 with 93.0% coverage.
  • Italian manual validation, rarer words. For words occurring 100–200 times in the corpus (558 pairs, 357 words), accuracy rose to 0.87 with 94.3% coverage.
  • Comparison with a manual oracle. The oracle contained 527 pairs, of which 437 were represented in MultiWordNet. The system produced 317 correct guesses, giving Precision 0.87 and Recall 0.72. Many of the remaining predictions were plausible new entries, such as dialectal, jargon or technical words.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics: Word sense disambiguation · Semantic role labeling

Cited by

19 papers of our library cite it (of the 140673 whose bibliographies we read); 5 of them have a page here, 14 not yet.

Cite

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

@inproceedings{cao2008combining,
  title = {{Combining Word Sense and Usage for Modeling Frame Semantics}},
  author = {Cao, Diego De and Croce, Danilo and Pennacchiotti, Marco and Basili, Roberto},
  booktitle = {Semantics in Text Processing. STEP 2008 Conference Proceedings},
  year = {2008},
  doi = {10.3115/1626481.1626489},
  url = {https://doi.org/10.3115/1626481.1626489}
}

References

24 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Agirre, E. and G. Rigau (1996). Word sense disambiguation using conceptual density. In Proceedings of COLING-96, Copenhagen, Denmark. find
  2. Baker, C. F., C. J. Fillmore, and J. B. Lowe (1998). The Berkeley FrameNet project. In Proceedings of COLING-ACL, Montreal, Canada. find our card
  3. Basili, R., M. Cammisa, and F. Zanzotto (2004). A semantic similarity measure for unsupervised semantic disambiguation. In Proceedings of LREC-04, Lisbon, Portugal. find
  4. Basili, R., D. D. Cao, P. Marocco, and M. Pennacchiotti (2007). Learning selectional preferences for entailment or paraphrasing rules. In Proceedings of RANLP 07. find
  5. Burchardt, A., K. Erk, and A. Frank (2005). A wordnet detour to framenet. In Proceedings of the GLDV 2005 GermaNet II Workshop, Bonn, Germany. find
  6. Burchardt, A., K. Erk, A. Frank, A. Kowalski, S. Pado, and M. Pinkal (2006). The salsa corpus: a german corpus resource for lexical semantics. In Proceedings of the 5th International Conference on Language Resources and Evaluation, Genova, Italy. find our card
  7. Burchardt, A., M. Pennacchiotti, S. Thater, and M. Pinkal (2008). Assessing the impact of frame semantics on textual entailment. Journal of Natural Language Engineering (to appear). find
  8. Clark, P., P. Harrison, J. Thompson, W. Murray, J. Hobbs, and C. Fellbaum (2007, June). On the Role of Lexical and World Knowledge in RTE3. In Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing, Prague, pp. 54–59. Association for Computational Linguistics. find
  9. Harris, Z. (1964). Distributional structure. In J. J. Katz and J. A. Fodor (Eds.), The Philosophy of Linguistics, New York. Oxford University Press. find
  10. Heyer, L., S. Kruglyak, and S. Yooseph (1999). Exploring expression data: Identification and analysis of coexpressed genes. Genome Research 9, 1106–1115. find
  11. Koehn, P. (2002). Europarl: A multilingual corpus for evaluation of machine translation. Draft. find
  12. De Cao, Croce, Pennacchiotti, and Basili Combining Word Sense and Usage for Modeling Frame Semantics Landauer, T. and S. Dumais (1997). A solution to plato’s problem: The latent semantic analysis theory of acquisition, induction and representation of knowledge. Psychological Review 104, 211–240. find
  13. Lin, D. and P. Pantel (2001). DIRT-discovery of inference rules from text. In Proceedings of the ACM Conference on Knowledge Discovery and Data Mining (KDD-01), San Francisco, CA. find
  14. Miller, G., R. Beckwith, C. Fellbaum, D. Gross, and K. Miller. (1990). An on-line lexical database. International Journal of Lexicography 13(4), 235–312. find
  15. Pado, S. (2007). Cross-Lingual Annotation Projection Models for Role-Semantic Information, Volume 21 of Saarbrücken Dissertations in Computational Linguistics and Language Technology. Saarland University. find
  16. Pado, S. and M. Lapata (2007). Dependency-based construction of semantic space models. Computational Linguistics 33(2), 161–199. find our card
  17. Padó, S., M. Pennacchiotti, and C. Sporleder (2008). Semantic role assignment for event nominalisations by leveraging verbal data. In Proceedings of COLING 2008, Manchester, UK. find
  18. Pianta, E., L. Bentivogli, and C. Girardi (2008). MultiWordNet: Developing an aligned multilingual database. In Proceedings of the 1st International Global WordNet Conference, Marrakech, Morocco, pp. 293–302. find
  19. Pitel, G. (2006). Using bilingual lsa for framenet annotation of french text from generic resources. In Workshop on Multilingual Semantic Annotation: Theory and Applications, SaarbrÃijcken, Germany. find
  20. Sahlgren, M. (2006). The Word-Space Model. Ph. D. thesis, Department of Linguistics, Stockholm University. find
  21. Schütze, H. (1993). Word space. In S. Hanson, J. Cowan, and C. Giles (Eds.), Advances in Neural Information Processing Systems 5. Morgan Kaufmann Publishers. find
  22. Shen, D. and M. Lapata (2007). Using semantic roles to improve question answering. In Proceedings of the Conference on Empirical Methods in Natural Language Processing and on Computational Natural Language Learning, Prague, pp. 12–21. find
  23. Subirats, C. and M. Petruck (2003). Surprise! Spanish FrameNet! In Proceedings of the Workshop on Frame Semantics at the XVII. International Congress of Linguists, Prague. find
  24. Tonelli, S. and E. Pianta (2008). Frame Information Transfer from English to Italian. In Proceedings of LREC 2008, Marrakech, Morocco. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/cao-2008-combining-word-sense-usage-modeling/