Advertisement

Using word association for syntactic disambiguation

  • R. Basili
  • M. T. Pazienza
  • P. Velardi
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 549)

Abstract

The study of word co-occurrences is common in linguistics but only recently the availability of on-line corpora and dictionaries made it possible to extensively collect word associations. Several papers published on this topic claim that their results are useful not only for lexicographers and linguists, but also for NLP, particularly for semantic and syntactic disambiguation in sentence analysis, as well as for lexical choice in generation. However such claims have not been convincingly proved so far. In this paper we argue that word associations derived through pure statistical analysis can hardly cope with the problem of syntactic disambiguation. It is shown that better performances are obtained by integrating wide-coverage techniques such as statistics with traditional NLP methods.

Keywords

Word Pair Word Association Computational Linguistics Prepositional Phrase Thematic Role 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [ACL 1990]
    Proceedings of ACL 90, Philadelfia, Pennsylvania, 1990.Google Scholar
  2. [Calzolari 1990]
    N. Calzolari, R. Bindi, Acquisition of Lexical Information from Corpus, Proc. of COLING 90, Helsinki 1990.Google Scholar
  3. [Church 1990]
    K.W.Church, P. Hanks, Word Association Norms, Mutual Information, and Lexicography, Computational Linguistics vol 16, n. 1, March 1990.Google Scholar
  4. [COL 1990]
    Proceedings of COLING 90, Helsinki 1990.Google Scholar
  5. [Fasolo 1990]
    M.Fasolo, L.Garbuio, N.Guarino, Comprensione di descrizioni di attivita' economico-produttive espresse in linguaggio naturale, Proc. of GULP Conference, Padova 1990.Google Scholar
  6. [Smadja 1989]
    F. A. Smadja, Lexical Co-occurrence: The Missing Link, Literary and Linguistic Computing, vol. 4, n.3, 1989.Google Scholar
  7. [Pazienza 1991]
    M.T. Pazienza, P.Velardi, Knowledge Acquisition for NLP: Tools and Methods, Proc. of The Int. Conference on Current Issues in Computational Linguistics, Penang, Malaysia, June, 1991.Google Scholar
  8. [Smadja 1989b]
    F. Smadja, Macrocoding the Lexicon with Co-occurrence Knowledge, in First Lexical Acquisition Workshop, August 1989, Detroit.Google Scholar
  9. [Tsutsumi 1991]
    T.Tsutsumi, Word sense disambiguation by examples, Proc. of The Int. Conference on Current Issues in Computational Linguistics, Penang, Malaysia, June, 1991.Google Scholar
  10. [Velardi 1991]
    P. Velardi, M.T. Pazienza, M. Fasolo, How to encode linguistic knowledge: a method for meaning representation and computer-aided acquisition, Computational Linguistics, 1991 (forthcoming).Google Scholar
  11. [Velardi 1991b]
    P. Velardi, Acquiring a semantic lexicon for natural language processing, in “Lexical Acquisition Using On-line Resources” U.Zernik editor, Lawrence Erlbaum ed., 1991 (in press).Google Scholar
  12. [Zernik 1990]
    U. Zernik, P. Jacobs, Tagging for Learning: Collecting Thematic Relations from Corpus, Proc. of COLING 90, Helsinki, August 1990.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • R. Basili
    • 1
  • M. T. Pazienza
    • 1
  • P. Velardi
    • 2
  1. 1.Dept. of Electr. EngineeringUniversity of Roma “Tor Vergata”RomaItaly
  2. 2.Institute of InformaticsUniversity of AnconaAnconaItaly

Personalised recommendations