Journal of KIISE (정보과학회 논문지)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지
DOI QR코드

DOI QR Code

Improvement of Korean Homograph Disambiguation using Korean Lexical Semantic Network (UWordMap)

한국어 어휘의미망(UWordMap)을 이용한 동형이의어 분별 개선

  • Received : 2015.07.22
  • Accepted : 2015.10.22
  • Published : 2016.01.15
⟨ Previous Next ⟩

Abstract

Disambiguation of homographs is an important job in Korean semantic processing and has been researched for long time. Recently, machine learning approaches have demonstrated good results in accuracy and speed. Other knowledge-based approaches are being researched for untrained words. This paper proposes a hybrid method based on the machine learning approach that uses a lexical semantic network. The use of a hybrid approach creates an additional corpus from subcategorization information and trains this additional corpus. A homograph tagging phase uses the hypernym of the homograph and an additional corpus. Experimentation with the Sejong Corpus and UWordMap demonstrates the hybrid method is to be effective with an increase in accuracy from 96.51% to 96.52%.

한국어처리 분야에서 동형이의어 분별은 의미처리를 위해서는 매우 중요하고 오랫동안 연구되어온 주제이다. 최근에 말뭉치를 학습하는 기계학습 방법이 정확률과 속도면에서 좋은 결과를 보이고 있으며, 미학습 어절을 처리하기 위해 어휘의미망을 이용한 지식기반 방법도 연구되고 있다. 본 논문은 말뭉치를 학습한 기계학습 방법에 어휘의미망과 함께 사용하는 방법을 제시한다. 이 방법의 기본 전략은 하위범주화 정보를 말뭉치화하여서 기존 말뭉치와 함께 학습시키고, 동형이의어 태깅 시점에서 분석 대상 명사의 상위어를 찾아서 학습정보와 같이 사용하는 것이다. 이 방법의 효과를 확인하기 위해 세종말뭉치와 UWordMap으로 실험을 하였으며, 정확률이 96.51%에서 96.52%로 미미하지만 상승하는 것을 확인하였다.

Keywords

Acknowledgement

Grant : Symbolic Approach 기반 인간모사형 자가 학습 지능 원천 기술 개발

Supported by : 정보통신기술진흥센터

References

  1. Joon-Choul Shin, C. Y. Ock, "A Korean Morphological Analyzer using a Pre-analyzed Partial Word-phrase Dictionary(in Korean)," Journal of KIISE : Software and Applications, Vol. 39, No. 5, pp. 415-424, May. 2012.
  2. Joon-Choul Shin, C. Y. Ock, "Korean Homograph Tagging model based on Sub-Word Conditional Probability(in Korean)," Journal of KIPS : Software and Data Enginering, Vol. 3, No. 10, pp. 407-420, Oct. 2014. https://doi.org/10.3745/KTSDE.2014.3.10.407
  3. Ji-Hui Im, H. O. Choe, C. Y. Ock, "Semantic Information Retrieval Based on User-Word Intelligent Network(in Korean)," Proc. of the Korea Contents Association Fall General Conference, Vol. 4, No. 2, pp. 547-550, 2006.
  4. Chulan Ren, Sehyeong Cho, "A Framework for WordNet-based Word Sense Disambiguation(in Korean)," Journal of the Korean Institute of Intelligent Systems, Vol. 23, No. 4, pp. 325-331, 2013. https://doi.org/10.5391/JKIIS.2013.23.4.325
  5. Dan Moldovan, Adrian Novischi, "Word sense disambiguation of WordNet glosses," Computer Speech and Language 18, pp. 301-317, 2004. https://doi.org/10.1016/j.csl.2004.05.007
  6. Minho Kim, Hyuk-Chul Kwon, "Word Sense Disambiguation using Semantic Relations in Korean WordNet (in Korean)," Journal of KIISE : Software and Applications, Vol. 38, No. 10, pp. 554-564, 2011.
  7. Sangwook Kang, and others, "Word Sense Disambiguation of Predicate using Sejong Electronic Dictionary and KorLex(in Korean)," Journal of KIISE : Computing Practices, Vol. 21, No. 7, pp. 500-505, 2015. https://doi.org/10.5626/KTCP.2015.21.7.500
  8. Young-jun Bae, C. Y. Ock, "Semantic Analysis of Korean Compound Noun using Lexical Semantic Network(U-WIN)(in Korean)," Journal of KIISE : Software and Applications, Vol. 40, No. 12, pp. 833-847, 2013.
  9. Joon-Choul Shin, Cheol-Young Ock, "A Study on Increasing Accuracy of Homograph Word Sense Disambiguation Using Korean Lexical Semantic Network(UWordMap)(in Korean)," Proc. of the Korea Computer Congress, pp. 817-819, 2015.
  10. Jung-Gil Cho, Kwang-Cheul Shin, "A Graph-based Word Sense Disambiguation Using Measures of Graph Connectivity(in Korean)," Journal of KIIT, Vol. 12, No. 6, pp. 143-151, Jun. 2014.
  11. Young-jun Bae, C. Y. Ock, "Intorduction to the Korean WordMap(UWordMap) and API(in Korean)," Proc. of 26th Annual Conference on Hamman and Cognitive Language Technology, Vol. 26, No. 1, pp. 27-31, 2014.

Cited by

  1. Comparing discrimination powers of text and citation-based context types pp.1588-2861, 2017, https://doi.org/10.1007/s11192-017-2566-9
  2. Effect of Word Sense Disambiguation on Neural Machine Translation: A Case Study in Korean vol.6, pp.2169-3536, 2018, https://doi.org/10.1109/ACCESS.2018.2851281