의미그룹을 이용한 단어 중의성 해소

Word Sense Disambiguation using Meaning Groups

  • 투고 : 2009.12.24
  • 심사 : 2010.03.25
  • 발행 : 2010.06.15

초록

본 연구에서는 기계 가독형 사전 정보를 이용하여 중의어 의미 태깅을 위한 학습에 사용될 양질의 의미부착 데이터를 자동 생성하여 중의어 태깅 정확도를 향상시키는 방법을 제안한다. 또한 중의어의 의미별로 특징있는 주변 단어를 원소로 갖는 의미그룹을 생성하고, 의미그룹내 중요 자질의 가중치 선정에 집중도 개념을 도입하여 태깅 정확도를 높인다. SENSEVAL-2 데이터를 사용하여 태깅 정확도를 평가한 결과 본 연구에서 제안한 방법이 기존의 방법보다 정확도를 향상시킨 것으로 나타났다.

This paper proposes the method that increases the accuracy for tagging word meaning by creating sense tagged data automatically using machine readable dictionaries. The concept of meaning group is applied here, where the meaning group for each meaning of a target word consists of neighbor words of the target word. To enhance the tagging accuracy, the notion of concentration is used for the weight of each word in a meaning group. The tagging result in SENSEVAL-2 data shows that accuracy of the proposed method is better than that of existing ones.

키워드

참고문헌

  1. H. Lee, D. H. Baek, H. C. Rim, "Word Sense Disambiguation using Classification Information," Journal of KIISE, vol.24 no.7, pp.779-789, 1997. (in korean)
  2. H. C. Seo, H. C. Rim, "Word Sense Disambiguation using Multiple Feature Decision Lists," Journal of KIISE, vol.30, no.7-8, pp.659-671, 2003.(in korean)
  3. D. S. Kim, J. W. Choe, "A Study on the Computational Model of Word Sense Disambiguation, based on Corpora and Experiments on Native Speaker's Intuition," Journal of KSCS, vol.17, no.4, pp.303-321, 2006. (in korean)
  4. H. Schutze, "Automatic Word Sense Discrimination," Proc. Computational Linguistics, 24(1): pp.97-123, 1998.
  5. S. I. Shin, K. S. Choi, "Automatic Clustering of Collocation for Detecting Practical Sense Boundary," Proc. ACL, 2004.
  6. M. Lesk, "Automatic Sense Disambiguation Using Machine Readable Dictionaries," Proc. SIGDOC Conference, pp.24-26, 1986.
  7. J. Hur, M. G. Jang, "Homonym Disambiguation based on Average Mutual Information," Proc. of the KIISE Korea Computer Congress, pp.159-166, 2005. (in korean)
  8. S. J. Lee, "A Related Keyword Group Extraction Method for Recommendation in Keyword Shop," M.S. Thesis, Soongsil University, Seoul Korea, 2003. (in korean)
  9. H. C. Seo, S. Z. Lee, H. C. Rim, and H. Lee, "KUNLP System using Classification Information Model at SENSEVAL-2," Proc. SENSEVAL-2, pp.147-150, 2001.