http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
딥러닝을 이용해 유의어의 변별 요소를 찾아보기 -여러 모델의 비교를 중심으로-
남유진 ( Nam You-jin ),여민수 ( Yeo Min-soo ) 한국어문학국제학술포럼 2021 Journal of Korean Culture Vol.55 No.-
This paper aims to find the distinctive factors of synonyms using various models of deep learning, what are the distinctive factors found in various models in common, and what it means to the research on synonyms. In the meantime, when we look at the flow of Korean lexical meaning research, from using the target word to create grammatically appropriate and non-grammatical sentences to distinguish their meaning, to using the corpus to find out what words the target word collocates with. In recent years, it has continued to identify the usage environment of target words from a large amount of data by applying a deep learning methodology. However, when introducing a methodology different from the previous one in this research flow, there are few discussions about why a new methodology should be introduced, and what the new methodology suggests and what its significance is. This paper examines what the semantic distinctive factors were identified in previous studies for '금방' and '방금', which have a synonyms relationship, and how the results of previous studies and the results of applying the deep learning methodology are the same and different. And by comparing the results, we will show the advantages of the deep learning method in the study of synonym distinction.