KIPS Transactions on Software and Data Engineering (정보처리학회논문지:소프트웨어 및 데이터공학)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

An Analysis of Korean Dependency Relation by Homograph Disambiguation

동형이의어 분별에 의한 한국어 의존관계 분석

  • 김홍순 (울산대학교 정보통신공학과) ;
  • 옥철영 (울산대학교 컴퓨터정보통신공학부)
  • Received : 2014.02.13
  • Accepted : 2014.05.13
  • Published : 2014.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

An analysis of dependency relation is a job that determines the governor and the dependent between words in sentence. The dependency relation of predicate is established by patterns and selectional restriction of subcategorization of the predicate. This paper proposes a method of analysis of Korean dependency relation using homograph predicate disambiguated in morphology analysis phase. The disambiguated homograph predicates has each different pattern. Especially reusing a stage transition training dictionary used during tagging POS and homograph, we propose a method of fixing the dependency relation of {noun+postposition, predicate}, and we analyze the accuracy and an effect of homograph for analysis of dependency relation. We used the Sejong Phrase Structured Corpus for experiment. We transformed the phrase structured corpus to dependency relation structure and tagged homograph. From the experiment, the accuracy of dependency relation by disambiguating homograph is 80.38%, the accuracy is increased by 0.42% compared with one of undisambiguated homograph. The Z-values in statistical hypothesis testing with significance level 1% is ${\mid}Z{\mid}=4.63{\geq}z_{0.01}=2.33$. So we can conclude that the homograph affects on analysis of dependency relation, and the stage transition training dictionary used in tagging POS and homograph affects 7.14% on the accuracy of dependency relation.

의존관계 분석은 문장의 어절 간에 의존소-지배소를 결정하는 작업이다. 용언은 문형 및 하위범주화 정보의 선택제약에 의해 다른 어절과의 의존관계를 형성한다. 본 논문은 형태소 분석 단계에서 동형이의어 분별된 용언의 문형을 이용하여 용언의 의존관계를 분석하는 방법을 제안한다. 특히, 형태소분석 단계에서 품사 및 동형이의어 태깅을 위해 사용하는 단계별 전이모델의 학습사전을 재활용하여 {명사+격조사, 용언} 간의 의존관계를 확정하는 방안을 제안하고 그의 정확률 및 영향을 분석한다. 동형이의어가 부착되고 의존관계로 변경된 21개의 세종구문분석말뭉치를 이용하여 실험한 결과, 동형이의어 분별된 의존관계 분석 정확률이 80.38%로, 동형이의어가 분별되지 않은 의존관계분석에 비해 0.42%의 정확률 향상이 있었으며, 유의수준 1%의 검정통계량 Z는 ${\mid}Z{\mid}=4.63{\geq}z_{0.01}=2.33$으로 동형이의어 분별이 의존관계 분석에 영향이 있음을 보였다. 또한, 단계별 전이모델이 의존관계 분석 정확률에 약 7.14% 영향을 미치는 것을 알 수 있었다.

Keywords

References

  1. J. Y. Oh and J. W. Cha, "Korean Dependency Parsing using Key Eojoel", Journal of KIISE : Software and Applications, Vol.40, No.10, pp.600-608, 2013.
  2. Y. U. Park and H. C. Kwon, "A Study of Parsing System Implementation Using Segmentation and Argument Information", Journal of Korea Multimedia Society, Vol.16, No.3, pp.366-374, 2013. https://doi.org/10.9717/kmms.2013.16.3.366
  3. E. K. Park and D. Y. Ra, "Processing Dependent Nouns Based on Chunking for Korean Syntactic Analysis", The Korean Society for Cognitive Science, Vol.17, No.2, pp.119-138, 2006.
  4. M. G. Jang, H. A. Lee, J. D. Park, and D. I. Park, "Korean Dependency Parser Using Subcategorization Information of Predicates", Journal of KIISE, pp.452-463, 1996.
  5. S. J. Lim, Y. T. Kim, and D. Y. Ra, "Korean Dependency Parsing Based on Machine Learning of Feature Weights", Journal of KIISE : Software and Applications, Vol.38, No.4, pp.214-223.
  6. Y. H. Lee and J. H. Lee, "Korean Dependency Parsing Using Online Learning", Journal of KIISE, pp.299-304, 2010.
  7. S. W. Jung, E. K. Park, D. Y. Ra, and J. T Yoon, "A Study on Korean Dependency Parser Using Case Relation and Mutual Information", Journal of KIISE, pp.450-455, 2001.
  8. Y. M. Park and J. Y. Seo, "Correction Method for Korean Dependency Parsing using Projectivity and Re-searching", Korean Journal of Cognitive Science, Vol.22, No.4, pp.429-447, 2011. https://doi.org/10.19066/cogsci.2011.22.4.005
  9. P. M. Ryu, T. S. Lee, J. H. Lee, and G. B. Lee, "Two-Phase Dependency Parser of Korean Using Predicate-Driven Constraint Propagation", Journal of KIISE, pp.923-926, 1996.
  10. S. S. Kim, S. B. Park, and S. J. Lee, "Analyzing Dependency of Korean Subordinate Clauses Using Support Vector Machine", Journal of KIISE, pp.148-155, 2006.
  11. G. E. Im, Y. G. Jung, and H. C. Kwon, "Implementation of Dependency Parser using Argument Information based on Korean WordNet", Journal of KIISE, pp.158-164, 2007.
  12. M. Y. Kim, S. J. Kang, and J. H. Lee, "Dependency Parsing by Chunks", Journal of KIISE, pp.327-329, 2000.
  13. S. W. Lee, "Cascaded Parsing Korean Sentences Using Grammatical Relations", pp.69-72. 2008. https://doi.org/10.3745/KIPSTB.2008.15-B.1.69
  14. S. H. Choi and H. R. Park, "Probabilistic Dependency Grammar Induction", The KIPS transactions, pp.513-515, 2003.
  15. S. S. Kim, S. B. Park, S. J. Lee, and S. Y. Park, "Analyzing dependency of Korean subordinate clauses using a composit kernel", Korean Journal of Cognitive Science, Vol.19, No.1, pp.1-15, 2008. https://doi.org/10.19066/cogsci.2008.19.1.001
  16. P. M. Ryu, J. H. Lee, and G. B. Lee, "Using Local Dependency for Dependecy Parser of Korean", The KIPS transactions, pp.464-468, 1996.
  17. J. H. Eun, M. W. Jeong, and G. B. Lee, "Korean Dependency Structure Analyzer based on Probabilistic Chart Parsing", Journal of KIISE, pp.105-111, 2005.
  18. S. B. Park and B. T. Zhang, "A Hybrid of Rule based Method and Memory based Learning for Korean Text Chunking", Journal of KIISE : Software and Applications, Vol.31, No.3, pp.369-378, 2004.
  19. L. K. Joo and J. H. Kim, "Implementing Korean Partial Parser based on Rules", pp.389-396, 2003. https://doi.org/10.3745/KIPSTB.2003.10B.4.389
  20. Y. M. Woo, Y. I. Song, S. Y. Park, and H. C. Rim, "Modification Distance Model using Headible Path Contexts for Korean Dependency Parsing", Journal of KIISE : Software and Applications, Vol.34, No.2, pp.140-149, 2007.
  21. J. C. Shin and C. Y. Ock, "A Stage Transition Model for Korean Part-of-Speech and Homograph Tagging", Journal of KIISE : Software and Applications, Vol.39, No.11, pp.889-901, 2012.
  22. Y. M. Park and J. Y. Seo, "SKA(Sogang Korean dependency Analyzer)", Competition of Korean Information Processing System, 2011.