Automatic Interpretation of Noun Compounds Using WordNet Similarity

What it is about

The paper interprets a novel binary noun compound, such as "paper submission", by finding the most similar training compound and copying its semantic relation. Similarity between the two nouns comes from WordNet::Similarity, with four measures compared. The relation set has 20 members, and the data is noun compounds from the Penn Treebank Wall Street Journal: 1,088 training and 1,081 test instances. The best measure, WUP, classifies 53.3% of the test compounds correctly. That is above the 43.0% majority baseline and at the level of the 52.3% inter-annotator agreement. The authors also expect the modifier and the head noun to carry different weight for different relations (for TIME, the modifier), and they test that with a weight parameter.

Ideas

• I1. Nearest-compound classifier in onto/src/wsd.rs with WordNet path similarity between modifier and head synsets, copying the relation of the nearest training compound. Measure: accuracy on EWT noun compounds against the majority-class baseline, two identical runs.
• I2. Relation-specific weight between modifier and head (the alpha of the paper) in en/src/explain.rs, fitted on a split. Measure: accuracy on the held-out split, with one weight versus a weight per relation.
• I3. Before measuring, count the share of binary compounds in the EWT test set. The paper reports 88.4% binary in the WSJ, and that share sets the scope. Measure: the share, reported with the first result.

For Mova

The chapter is 18. Compounding, conversion and the minor processes. The code is onto/src/wsd.rs and en/src/explain.rs. The WSJ material is licensed Penn Treebank data, so it is a test only, and the paper is NC; the 20-relation set is a design source, not something to copy into a release.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.1007/11562214_82

@inproceedings{kim2005automatic,
  title = {{Automatic Interpretation of Noun Compounds Using WordNet Similarity}},
  author = {Kim, Su Nam and Baldwin, Timothy},
  booktitle = {Second International Joint Conference on Natural Language Processing: Full Papers},
  year = {2005},
  doi = {10.1007/11562214\_82},
  url = {https://doi.org/10.1007/11562214_82}
}

References

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

1. Cao, Y., Li, H.: Base noun phrase translation using web data and the em algorithm. In: COLING2002. (2002) find
2. Baldwin, T., Tanaka, T.: Translation by machine of compound nominals: Getting it right. In: ACL2004-MWE, Barcelona, Spain (2004) 24–31 ’correct.percent’ ’incorrect.percent’ THRESHOLD Similarity=0.57 Error 0.4 0.6 0.8 (similarity) find
3. Lauer, M.: Designing Statistical Language Learners: Experiments on Noun Compounds. PhD thesis, Macquarie University (1995) find
4. Lapata, M.: The disambiguation of nominalizations. Comput. Linguist. 28 (2002) 357–388 find
5. Finin, T.W.: The semantic interpretation of compound nominals. PhD thesis, University of Illinois, Urbana, Illinois, USA (1980) find
6. Vanderwende, L.: Algorithm for automatic interpretation of noun sequences. In: Proceedings of the 15th conference on Computational linguistics. (1994) 782–788 find
7. Rosario, B., Marti, H.: Classifying the semantic relations in noun compounds via a domain-specific lexical hierarchy. In: Proceedings of the 2001 Conference on Empirical Methods in Natural Language Processing. (2001) 82–90 find our card
8. Moldovan, D., Badulescu, A., Tatu, M., Antohe, D., Girju, R.: Models for the semantic classification of noun phrases. HLT-NAACL 2004: Workshop on Computational Lexical Semantics (2004) 60–67 find
9. Fan, J., Barker, K., Porter, B.W.: The knowledge required to interpret noun compounds. In: Seventh International Joint Conference on Artificial Intelligence. (2003) 1483–1485 find
10. Barker, K., Szpakowicz, S.: Semi-automatic recognition of noun modifier relationships. In: Proceedings of the 17th international conference on Computational linguistics. (1998) 96– 102 find
11. Artiles, J., Penas, A., Verdejo, F.: Word sense disambiguation based on term to term similarity in a context space. In: Senseval-3: Third International Workshop on the Evaluation of Systems for the Semantic Analysis of Text. (2004) 58–63 find
12. Patwardhan, S., Banerjee, S., Pedersen, T.: Using measures of semantic relatedness for word sense disambiguation. In: Proceedings of the Fourth International Conference on Intelligent Text Processing and Computational Linguistics. (2003) find
13. Resnik, P.: Disambiguating noun groupings with respect to wordnet senses. In: Proceedings of the 3rd Workship on Very Large Corpus. (1995) 77–98 find
14. Wu, Z., Palmer, M.: Verb semantics and lexical selection. In: 32nd. Annual Meeting of the Association for Computational Linguistics. (1994) 133 –138 find
15. Fellbaum, C., ed.: WordNet: An Electronic Lexical Database. MIT Press, Cambridge, USA (1998) find
16. Leacock, C., Chodorow, N.: Combining local context and wordnet similarity for word sense identification. [15] find
17. Jiang, J., Conrath, D.: Semantic similarity based on corpus statistics and lexical taxonomy. In: Proceedings on International Conference on Research in Computational Linguistics. (1998) 19–33 find
18. Lin, D.: An information-theoretic definition of similarity. In: Proceedings of the International Conference on Machine Learning. (1998) find
19. Banerjee, S., Pedersen, T.: Extended gloss overlaps as a measure of semantic relatedness. In: Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence. (2003) 805–810 find
20. Levi, J.: The syntax and semantics of complex nominals. In: New York:Academic Press. (1979) find
21. Downing, P.: On the creation and use of English compound nouns. Language 53 (1977) 810–42 find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2005-kim-automatic-interpretation-noun-compounds-wordnet/