slider01 slider02 slider03

DOI: http://dx.doi.org/10.17959/sppm.2023.29.3.329

PDF: 본문파일


음소배열정보 기반 한국어 고유어, 한자어, 차용어의 머신러닝 분류

박선우 (계명대학교)

Abstract

The purpose of this study is to test models that automatically classify Korean nouns 
into native Korean, Sino-Korean, and loanwords by applying a machine learning 
model, naïve Bayes classification. In this study, 500 native Korean words, Sino- 
Korean words, and loanwords were collected, and after romanizing and decomposing 
them into bigram and trigram lists, the bigrams and trigrams were entered into the 
naïve Bayes classifier. We tested models with and without syllable boundaries, and 
found that both the bigram and trigram models were over 80% accurate. Contrary to 
the  expectation  that  the  performance  of  the  models  would  improve  as  more 
information about Korean phonotactics was included in the training and validation 
data, the difference in performance between the bigram and trigram models was not 
significant. The model that included syllable boundaries in the phoneme sequence 
information had slightly higher accuracy than the model without syllable boundary 
information.  When  comparing  the  classification  results  of  all  five  models,  the 
accuracy of the bigram model with syllable boundaries was 83.55%, which was the 
best. For now, we have modified the model to consider only phoneme sequence 
information and syllable boundaries, but it is expected that the accuracy of the model 
can be improved by training the model while excluding bigrams and trigrams, which 
occur in similar proportions in all categories, and by increasing the size of the data. 

Keywords
phonotactics, native Korean, Sino-Korean, loanword, machine learning, Naïve Bayes classification, bigram model, trigram model 
번호 제목 글쓴이 날짜 조회 수
공지 [음성음운형태론연구] 온라인 논문 투고 안내 (2023년 1월 14일 수정) Manager 2016.09.02 30500
공지 [음성음운형태론연구] 논문 투고시 유의사항 (2023년 1월 14일 수정) Manager 2013.04.27 39041
687 음성음운형태론연구 29집 3호 목록 홍보이사_2 2024.01.31 85
686 음성음운형태론연구 29집 3호 Hong, Soonhyun 홍보이사_2 2024.01.31 59
685 음성음운형태론연구 29집 3호 Kuwamoto, Yuji 홍보이사_2 2024.01.31 67
684 음성음운형태론연구 29집 3호 Lee, Joo Kyeong 홍보이사_2 2024.01.31 53
683 음성음운형태론연구 29집 3호 Yu, Hye Jeong 홍보이사_2 2024.01.31 64
682 음성음운형태론연구 29집 3호 Ahn, Miyeon 홍보이사_2 2024.01.31 42
681 음성음운형태론연구 29집 3호 Park, Shinae 홍보이사_2 2024.01.31 47
» 음성음운형태론연구 29집 3호 박선우 홍보이사_2 2024.01.31 54
679 음성음운형태론연구 29집 2호 목록 홍보이사_2 2023.09.09 222
678 음성음운형태론연구 29집 2호 Whang, James & Yazawa, Kakeru 홍보이사_2 2023.09.09 737
677 음성음운형태론연구 29집 2호 Tilsen, Sam 홍보이사_2 2023.09.09 193
676 음성음운형태론연구 29집 2호 Hong, Soonhyun 홍보이사_2 2023.09.09 132
675 음성음운형태론연구 29집 2호 Lee, Yong-cheol 홍보이사_2 2023.09.09 109
674 음성음운형태론연구 29집 2호 Sung, Eunkyung 홍보이사_2 2023.09.09 98
673 음성음운형태론연구 29집 1호 목록 홍보이사_2 2023.05.28 241
672 음성음운형태론연구 29집 1호 이주희 홍보이사_2 2023.05.28 197
671 음성음운형태론연구 29집 1호 Hye Joeng Yu 홍보이사_2 2023.05.28 152
670 음성음운형태론연구 29집 1호 Hyebae Yoo 홍보이사_2 2023.05.28 147
669 음성음운형태론연구 29집 1호 Miyeon Ahn 홍보이사_2 2023.05.28 125
668 음성음운형태론연구 29집 1호 Jae-Hyun Sung, Tae-Jin Yoon, Soohyun Kwon, Gwanhi Yun 홍보이사_2 2023.05.28 126