Cascaded Grammatical Relation-Driven Parsing Using Support Vector Machines

  • Songwook Lee
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4188)


This study aims to identify dependency structure in Korean sentences with the cascaded chunking strategy. In the first stages of the cascade, we find chunks of NP and guess grammatical relations (GRs) using Support Vector Machine (SVM) classifiers for every possible modifier-head pairs of chunks in terms of GR categories as subject, object, complement, adverbial, and etc. In the next stage, we filter out incorrect modifier-head relations in each cascade for its corresponding GR using the SVM classifiers and the characteristics of the Korean language such as distance, no-crossing and case property. Through an experiment with a tree and GR tagged corpus for training the proposed parser, we achieved an overall accuracy of 85.7% on average.


Support Vector Machine Noun Phrase Dependency Structure Correct Relation Training Support Vector Machine 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Brants, T., Skut, W., Krenn, B.: Tagging grammatical functions. In: Proc. of the 2nd Conference on EMNLP, pp. 64–74 (1997)Google Scholar
  2. 2.
    Argamon, S., Dagan, I., Krymolowski, Y.: A memory-based approach to learning shallow natural language patterns. In: Proc. of the 36th Annual Meeting of the ACL, pp. 67–73 (1998)Google Scholar
  3. 3.
    Buchholz, S., Veenstra, J., Daelemans, W.: Cascaded GR assignment. In: Proc. of the Joint Conference on EMNLP and Very Large Corpora, pp. 239–246 (1999)Google Scholar
  4. 4.
    Blaheta, D., Charniak, E.: Assigning function tags to parsed text. In: Proc. of the 1st Conference of the NAACL, pp. 234–240 (2000)Google Scholar
  5. 5.
    Carroll, J., Briscoe, E.: High precision extraction of grammatical relations. In: Proc. of the 19th International Conference on Computational Linguistics, pp. 134–140 (2002)Google Scholar
  6. 6.
    Joachims, T.: Text Categorization with Support Vector Machines: Learning with Many Relevant Features. In: Proc. of European Conference on Machine Learning, pp. 137–142 (1998)Google Scholar
  7. 7.
    Lee, S., Seo, J., Jang, T.Y.: Analysis of the grammatical functions between adnoun and NPs in Korean using Support Vector Machines. Natural Language Engineering 9(3), 269–280 (2003)Google Scholar
  8. 8.
    Lee, K.J., Kim, J.H., Kim, G.C.: An Efficient Parsing of Korean Sentence Using Restricted Phrase Structure Grammar. Computer Processing of Oriental Languages 12(1), 49–62 (1997)Google Scholar
  9. 9.
    Viterbi, A.J.: Error bounds for convolution codes and an asymptotically optimal decoding algorithm. IEEE trans. on Information Theory 12, 260–269 (1967)CrossRefGoogle Scholar
  10. 10.
    Vapnik, V.N.: The Nature of Statistical Learning Theory. Springer, New York (1995)MATHGoogle Scholar
  11. 11.
    Lee, K.J., Kim, J.H., Choi, K.S., Kim, G.C.: Korean syntactic tagset for building a tree annotated corpus. Korean Journal of Cognitive Science 7(4), 7–24 (1996)Google Scholar
  12. 12.
    van Rijsbergen, C.J.: Information Retrieval. Buttersworth, London (1979)Google Scholar
  13. 13.
    Lee, S.: A statistical model for identifying grammatical relations in Korean sentences. IEICE transactions on Information and Systems E87-D(12), 2863–2871 (2004)Google Scholar
  14. 14.
    Lee, S., Seo, J.: Grammatical relations identification of Korean parsed texts using support vector machines. In: Sojka, P., Kopeček, I., Pala, K. (eds.) TSD 2004. LNCS (LNAI), vol. 3206, pp. 121–128. Springer, Heidelberg (2004)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Songwook Lee
    • 1
  1. 1.Division of Computer and Information EngineeringDongseo UniversityBusanKorea

Personalised recommendations