[논문]딥러닝을 이용한 한국어 Head-Tail 토큰화 기법과 품사 태깅

김정민; 강승식; 김혁만

doi:10.14372/iemek.2022.17.4.199

딥러닝을 이용한 한국어 Head-Tail 토큰화 기법과 품사 태깅
Korean Head-Tail Tokenization and Part-of-Speech Tagging by using Deep Learning 원문보기

대한임베디드공학회논문지 = IEMEK Journal of embedded systems and applications, v.17 no.4, 2022년, pp.199 - 208

김정민 (Kookmin University) , 강승식 (Kookmin University) , 김혁만 (Kookmin University)

Abstract ▼ AI-Helper

Korean is an agglutinative language, and one or more morphemes are combined to form a single word. Part-of-speech tagging method separates each morpheme from a word and attaches a part-of-speech tag. In this study, we propose a new Korean part-of-speech tagging method based on the Head-Tail tokenization technique that divides a word into a lexical morpheme part and a grammatical morpheme part without decomposing compound words. In this method, the Head-Tail is divided by the syllable boundary without restoring irregular deformation or abbreviated syllables. Korean part-of-speech tagger was implemented using the Head-Tail tokenization and deep learning technique. In order to solve the problem that a large number of complex tags are generated due to the segmented tags and the tagging accuracy is low, we reduced the number of tags to a complex tag composed of large classification tags, and as a result, we improved the tagging accuracy. The performance of the Head-Tail part-of-speech tagger was experimented by using BERT, syllable bigram, and subword bigram embedding, and both syllable bigram and subword bigram embedding showed improvement in performance compared to general BERT. Part-of-speech tagging was performed by integrating the Head-Tail tokenization model and the simplified part-of-speech tagging model, achieving 98.99% word unit accuracy and 99.08% token unit accuracy. As a result of the experiment, it was found that the performance of part-of-speech tagging improved when the maximum token length was limited to twice the number of words.

주제어

표/그림 (18)

표 표 1. KCC150 형태 분석 말뭉치 예제 Table 1. A Sample of KCC150 Morphological Analysis Corpus
표 표 2. Head-Tail 말뭉치 예제 Table 2. A Sample of Head-Tail Corpus
그림 그림 1. Head-Tail 형태소 분석 과정 Fig. 1. Head-Tail Morphological Analysis Process
그림 그림 2. Head-Tail 토크나이저의 입출력 Fig. 2. Head-Tail Tokenizer Input & Output
그림 그림 3. BiLSTM을 이용한 Head-Tail 토크나이저 모델 Fig. 3. Head-Tail Tokenizer Model by using BiLSTM
그림 그림 4. Head-Tail 품사 태깅 모델 Fig. 4. Head-Tail Part-of-Speech Tagging Model
그림 그림 5. Sequence Output Concat Fig. 5. Sequence Output Concat
그림 그림 6. One Output Embedding Concat (Replace) Fig. 6. One Output Embedding Concat (Replace)
표 표 3. Subword 토큰 단위의 BIO 데이터셋 예 Table 3. BIO Dataset Example by Subword Tokens
표 표 4. 토크나이저 모델별 정확도 Table 4. Accuracy of Tokenizer Models
표 표 5. 품사 태깅 모델별 정확도 Table 5. Accuracy of Part-of-Speech Tagging Model
표 표 6. 토크나이저 태거 통합 정확도 Table 6. Merge Accuracy of Tokenizer & Tagger
표 표 7. 에러가 발생한 단일/복합 태그의 분포 Table 7. Distribution of Single/Complex Tags with Errors
표 표 8. 오분류 태그의 대분류 태그 분포 Table 8. Large Category Tag Distribution of Misclassified Tags
표 표 9. 태그 간소화후 품사태거 모델별 정확도 Table 9. Accuracy by Model after Tag Simplification
표 표 10. 태그 간소화 후 토크나이저 태거 통합 정확도 Table 10. Accuracy of Tokenizer and Tagger after Tag Simplification
표 표 11. 품사 태깅 모델의 성능 평가 Table 11. Performance Evaluation of Part-of-Speech Tagging Model
표 표 12. 형태소 분석기별 분류 실험 결과 Table 12. Classification Results by Morphological Analyzer

참고문헌 (28)

D. H. Park, W. S. Choi, H. J. Kim, S. L. Lee, "Web Document Classification Based on Hangeul Morpheme and Keyword Analyses", Journal of Korean Information Processing Society D, Vol. 19, No. 4, pp. 263-270, 2012 (in Korean).
W. Cho, G. Shin, W. Lee, S. Son, H. Song, J. Lee, H. Lee, S. Jo, "KoELECTRA-Based Named Entity Recognition Using Korean Morphological Analyzers", Proceedings of 2021 Korea Computer Congress, pp. 1897-1899, 2021 (in Korean).
Y. Ha, J. Cheon, I. Wang, M. Park, G. Woo, "A Filtering Method of Malicious Comments Through Morpheme Analysis", Journal of Korea Contents Association, Vol. 21, No. 9, pp. 750-761, 2021 (in Korean).

원문보기 상세보기
S. S. Kang, "Analysis of Korean Irregular Verbs by using Syllable Characteristics", Proceedings of 5th Hangul and Korean Information Processing, pp. 385-394, 1993 (in Korean).
S. S. Kang, "Multi-level Morphology and Morphological Analysis Model for Korean", Proceedings of 6th Hangul and Korean Information Processing, pp. 140-145, 1994 (in Korean).
J. S. Lee, "Three-Step Probabilistic Model for Korean Morphological Analysis", Journal of KIISE, Vol. 38, No. 5, pp. 257-268, 2011 (in Korean).
D. Bahdanau, K. Cho, Y. Bengio, "Neural Machine Translation by Jointly Learning to Align and Translate", ICLR 2015, arXiv:1409.0473, 2015.
A. Vaswan, N. Shazeer., N. Parmar., J. Uszkoreit., L. Jones.,N. Aidan. Gomez, L. Kaiser., L. Polosukhin., "Attention Is All You Need", Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS), arXiv:1706.03762v5, 2017.
J. Devlin., M. W. Chang., K. Lee., K. Toutanova., "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding", preprint arXiv:1810.04805, 2018.
H. Y. Lee, S. S. Kang, "Bi-LSTM-CRF and Syllable Embedding for Automatic Spacing of Korean Sentences", Proceedings of 30th Hangul and Korean Information Processing, pp. 605-607, 2018 (in Korean).
K. S. Shim, "Automatic Word Spacing based on Conditional Random Fields", Korean Journal of Cognitive Science, Vol. 22, No. 2, pp. 217-233, 2011 (in Korean).

원문보기 상세보기
H. Y. Lee, S. S. Kang, "Compound Noun Decomposition by using Syllable-based Embedding and Deep Learning", Journal of Smart Media, Vol. 8, No. 2, pp. 74-79, 2019 (in Korean).
J. H. Shin, H. R. Park, "A Statistical Model for Korean Text Segmentation by using Syllable-level Bigrams", Proceedings of 9th Hangul and Korean Information Processing, pp. 255-260, 1997 (in Korean).
S. S. Kang, "Automatic Correction of Word-spacing Errors by using Syllable Bigram", Journal of Speech Science, Vol. 8, No. 2, pp. 83-90, 2001 (in Korean).
D. H. Lim, Y. J. Chun, H. J. Kim, S. S. Kang, "Word Segmentation by using Extended Syllable Bigram", Proceedings of 17th Hangul and Korean Information Processing, pp. 189-193, 2005 (in Korean).
T. Kudo.., "Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates", Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, pp. 66-75, 2018.
T. Kudo, J. Richardson., "SentencePiece: A Simple and Language Independent Subword Tokenizer nd Detokenizer for Neural Text Processing", Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System, pp. 66-71. 2018.
J. Lafferty, A. McCallum, F. C. N. Pereira, "Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data", Proceedings of the 18th International Conference on Machine Learning 2001, pp. 282-289, 2001.
T. Brants., "TnT - A Statistical Part-of-Speech Tagger", Proceedings of the 6th Applied Natural Language Processing Conference, pp. 224-231, 2000.
J. Youn, J. Lee, "A Deep Learning-based Two-Steps Pipeline Model for Korean Morphological Analysis and Part-of-Speech Tagging", Journal of KIISE, Vol. 48, No. 4, pp. 444-452, 2021.
J. Li, E. H. Lee, J. H. Lee, "Sequence-to-sequence based Morphological Analysis and Part-Of-Speech Tagging for Korean Language with Convolutional Features", Journal of KIISE, Vol. 44, No. 1, pp. 57-62, 2017 (in Korean).

원문보기 상세보기
Y. Choi, K. Lee, "Performance Analysis of Korean Morphological Analyzer based on Transformer and BERT", Journal of KIISE, Vol. 47, No. 8, pp. 730-741, 2020 (in Korean).

상세보기
J. M. Kim, H. J. Seo, S. S. Kang, "Korean Head-Tail POS-Tagger by using Transformer", Proceedings of 33th Hangul and Korean Information Processing, pp. 544-547, 2021 (in Korean).
D. Kim, Y. Bae, C. Ock, H. Choi, C. Kim, "Korean POS and Homonym Tagging System using HMM", Proceedings of 20th Hangul and Korean Information Processing, pp. 12-16, 2008 (in Korean).
J. H. Kim, "Korean Part-of-Speech Tagging using a Weighted Network", Journal of KIISE, Vol. 25, No. 6, pp. 951-959, 1998 (in Korean).
K. S. Shim, "Syllable-based POS Tagging without Korean Morphological Analysis", Korean Journal of Cognitive Science, Vol. 22, No. 3, pp. 327-345, 2011 (in Korean).

원문보기 상세보기
J. M. Kim, H. J. Seo, S. S. Kang, "Korean Head-Tail POS-Tagger by using Transformer", Journal of Smart Media, Vol. 11, No. 5, pp. 17-25, 2022 (in Korean).
D. Cho, "Hate Speech Detection by using Word Embedding and Deep Learning Model", M.S. Thesis, Kookmin University, 2021 (in Korean).

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증