쉼표의 자동분류에 따른 중국어 장문분할 = Segmentation of Long Chinese Sentences using Comma Classification

국문 초록 (Abstract)

입력문장이 길어질수록 구문분석의 정확률은 크게 낮아진다. 따라서 긴 문장의 구문분석 정확률을 높이기 위해 장문분할 방법들이 많이 연구되었다. 중국어는 고립어로서 자연언어처리에 도움을 줄 수 있는 굴절이나 어미정보가 없는 대신 쉼표를 비교적 많이, 또 정확히 사용하고 있어서 이러한 쉼표사용이 장문분할에 도움을 줄 수 있다. 본 논문에서는 중국어 문장에서 쉼표 주변의 문맥을 파악하여 해당 쉼표위치에 문장분할이 가능한지 Support Vector Machine을 이용해 판단하고자 한다. 쉼표의 분류의 정확률이 87.1%에 이르고, 이 분할모델을 적용한 후 구문분석한 결과, 의존트리의 정확률이 5.6% 증가했다.

다국어 초록 (Multilingual Abstract)

The longer the input sentences, the worse the parsing results. To improve the parsing performance, many methods about long sentence segmentation have been reserarched. As an isolating language, Chinese sentence has fewer cues for sentence segmentation. However, the average frequency of comma usage in Chinese is higher than that of other languages. The syntactic information that the comma conveys can play an important role in long sentence segmentation of Chinese languages. This paper proposes a method for classifying commas in Chinese sentences according to the context where the comma occurs. Then, sentences are segmented using the classification result. The experimental results show that the accuracy of the comma classification reaches 87.1%, and with our segmentation model, the dependency parsing accuracy of our parser is improved by 5.6%.

목차 (Table of Contents)

• 요약
• Abstract
• 1. 서론
• 2. 기존연구
• 3. 구문분석을 위한 쉼표의 분류
• 4. 쉼표의 분류를 위한 자질 추출
• 5. 실험
• 6. 기존연구와의 비교 및 결론
• 참고문헌
• 저자소개

참고문헌 (Reference)

• 1 Geoffrey Nunberg, "the linguistics of punctuation" 18 : 1990.
• 2 Shui-fang Lin, "study and application of punctuation" People’s Publisher 2000.
• 3 B. Say, "current approaches to punctuation in computational linguistics" 30 (30): 457-469, 1997
• 4 V.J. Leffa, "clause processing in complex sentences" 937-943, 1998.
• 5 B. Jones, "What’s the point? A(computational) theory of punctuation" 1996.
• 6 D.M.Bikel, "Two statistical parsing models applied to the Chinese Treebank" 1-6, 2000.
• 7 B. Jones, "Towards testing the syntax of punctuation" 363-365, 1996.
• 8 V N. Vapnik, "The nature of statistical learning theory" Springer-Verlag 1995
• 9 N. Xue, "The bracketing Guidelines for the Penn Chinese Treebank University of Pennsylvania" 2000.
• 10 H. Yamada, "Statistical Dependency Analysis with Support Vector Machines" 195-206, 2003
• 11 M.Y.Kim, "Resolving ambiguity in Inter-chunk dependency parsing" 2001.
• 12 T.Joachims, "Making large-Scale SVM Learning Practical. Advances in Kernel Methods - Support Vector Learning" MIT-Press 1999.
• 13 S. Kim, "Learning-based intrasentence segmentation for efficient translation of long sentences" 16 (16): 151-174, 2001.
• 14 X. Carreras, "Learning and inference for clause identification" 35-47, 2002.
• 15 Roger Levy, "Is it harder to parse Chinese, or the Chinese Treebank?" 439-446, 2003.
• 16 R.Levy, "Is it Harder to Parse Chinese, or the Chinese Treebank?" 2003.
• 17 E.F.T.K. Sang, "Introduction to the CoNLL-2001 shared task:clause identification" 53-57, 2001.
• 18 M. J. Collins, "Head-driven Statistical Models for Natural Language Parsing" 1999.
• 19 B. Jones, "Exploring the role of punctuation in parsing natural text" 421-425, 1994.
• 20 Briscoe,E, "Developing and evaluating a probabilistic LR parser of part-of-speech and punctuation labels" 48-58, 1995.
• 21 M. Bayparktar, "An analysis of English punctuation the special case of comma" 3 (3): 33-57, 1998.
• 22 P.L. Shiuan, "A divide-and- conquer strategy for parsing" 57-66, 1996
• 23 P.L. Shiuan, "A divide-and- conquer strategy for parsing" 57-66, 1996
• 24 R.L. Hill, "A comma in parsing: A study into the influence of punctuation (commas) on contextually isolated" M.Phil disseration 1996.