Earticle

본 연구에서는 언어 처리 연구의 일환으로 텍스트 마이닝 기법을 활용하여 ‘한・일 어 대조연구(日韓対照研究)’의 동향을 파악하고 그 결과를 분석하였다. 구체적으로는 한일 양 언어의 대조연구가 언제부터 발표되었으며, 주로 어떤 분야가 연구되고 있는가 를 포함해 지금까지의 연구 성과와 동향을 파악하는데 주안점을 두었다. 이를 위한 연 구 방법으로 ‘Google Scholar’ 검색툴을 이용하여 “日韓対照”라는 키워드로 논문을 검 출하였다. 그 결과, 1980년부터 2022년까지 발행된 논문 783편을 수집하였고, 이들 논문 타이틀을 대상으로 KH Coder를 활용한 텍스트 마이닝을 시도하였다. 분석 결과를 아래 에 요약한다. [1] 텍스트 마이닝을 활용하여 논문의 타이틀에 출현하는 상위 빈도어(4회 이상 출 현)를 추출하였다. 대조연구의 분야로는 ‘문법・표현’ 및 ‘담화・언어행동’이 활 발하나 ‘어휘・품사’나 ‘음성・음운’을 비롯해 그 밖의 분야에서도 폭넓게 진행 되고 있다. [2] 추출어 간의 상관관계를 확인하기 위해 공기네트워크를 실행하여 상위 100개의 주제어가 논문 타이틀에서 어떻게 연결되어 있는가를 확인하였다. 그 결과, [日 本語] [韓国語] [動詞] [談話] [行動] [形式] 등이 큰 클러스터를 이루고 있음을 확 인하였다. [3] 발행연도를 외부변수로 설정하여 지난 40여년 사이의 대조연구의 추이와 연대별 특징적인 추출어를 확인하였다. ‘문법・담화’분야에서는 초기의 [受身・使役]의 연구에서 [テンス・アスペクト]의 연구를 거쳐 2000년대에 들어 [談話] 연구가 출현하였고, 최근에는 [場面・発話]를 중시한 [コミュニケーション]의 연구로 이 어지고 있음을 확인하였다.

In this study, as part of language processing research, text mining techniques were used to identify trends in ‘Japanese-Korean contrastive research’ and analyze the results. Specifically, the focus was on identifying research results and trends to date, including when contrastive studies on both Korean and Japanese languages w ere published and what areas are mainly being studied. As a research method for this, we first used Google Scholar's search tool to detect papers matching the keyword “Japanese-Korean contrastive Research”. We collected 783 papers published from 1980 to 2022, and attempted text mining using KH Coder on these paper titles. The analysis results of this study are summarized below. [1] Using text mining, we extracted high-frequency words (appearing 4 or more times) that appear in the title of the paper. The fields of contrastive research are active in ‘Grammar/Expression’ and ‘Discourse/Verbal behavior’, but a wide range of research is also being conducted in ‘Vocabulary/Parts of speech’, ‘Phonetics/Phonology’, and other fields. [2] To check the correlation between extracted words, we ran co-occurrence network to check how the top 100 keywords are connected in the paper title. As a result, it was confirmed that [Korean language] [Japanese language] [verbs] [form] [discourse] [verbal behavior] etc. formed a large cluster. [3] By setting the year of publication as an external variable, we confirmed contrastive research trends over the past 40 years and extracted characteristics by era. In the field of ‘grammar/discourse’, from the early [causative/passive] [particle] research to [tense/aspect] research, research on [discourse] emerged in the 2000s. Recently, research on [communication] that emphasizes [situation/utterance] has continued.

최근 제4차 산업 혁명의 시대로 접어들면서 인공 지능(AI)이나 빅 데이터 등이 이슈가 되고 있다. 이러한 변화 속에서 인문학 분야에서도 데이터 마이닝 등을 활용 한 연구가 시도되고 있다. 본 논문에서는 일본어학 분야에서 새로운 연구방법으로 대두되고 있는 ‘텍스트 마이닝’ 기법을 활용하여 ‘일본어 조사(助詞)’ 연구를 시도하였다. 따라서 본 연구는 일본어의 조사 연구에 있어서 ‘계량적・정량적・통계적 분석의 가능성’을 시도한 데 에 의의가 있다. 분석 대상은 일본어 조사 ‘X {まで(made)}’를 중심으로 ‘X {さえ(sae)／も(mo)}’와 비교・대조 고찰하였다. 분석 엔진으로는 樋口(HiGuchi)가 개발한 ‘KH Coder’를 활 용하였고, 이 툴은 조사 등의 기능어도 분석이 가능하다. 이 KH Coder를 통해 ‘X {まで／さえ／も}’ 구문의 앞에 오는 전접 요소의 ‘빈출 어’와 ‘공기 네트워크’, ‘대응분석’과 ‘특징어’를 확인할 수 있었다. 이는 기존에 없 었던 새로운 연구방법을 도입하여 일본어 조사의 분석을 시도한 것으로 그 방법의 독창성과 참신성이 인정된다. 또 실제 코퍼스 자료를 사용함으로써 실증적인 연구로 서도 인정할 수 있다. 본 연구의 텍스트 마이닝 기법은, 일본어학 연구 특히 일본어의 조사에 대한 접 근방법의 폭을 넓히고 있다. 시드 데이터 분석의 접근과 해석 방식, 키워드 도출 및 시각화 분석을 통해 명료성과 객관성을 확보하여 학제 간 연구의 자료로 활용될 것 으로 기대된다.

Recently, in the era of the “Fourth Industrial Revolution”, “artificial intelligence(AI)” and “big data” have created certain issues. In such scenario, research utilizing “data mining” has been attempted in the field of humanities as well. In the current research, attempts have been made to study “Japanese particles” utilizing the “text mining” technique that emerged as a new research method in the field of Japanese language studies. It would be prudent to try “the possibility of quantitative, quantitative, and statistical analysis” in this study of the Japanese particles. The analysis target was compared and contrasted with “X {sae(even) / mo(also)}”, focusing on the Japanese particle “X {made(until)}”. The analysis engine used “KH Coder” developed by HiGuchi(樋口); however, this tool can also analyze function words such as particles. Using the KH Coder, we were able to confirm the “frequent words” and “co-occurrence network”, “correspondence analysis” and “characteristic words” of “X {made / sae / mo}” syntax which are commonly not understood through the conventional research methods. We intend to use the text mining technique more as “objective” data to analyze “keyword” derivation and “visualization” by expanding the scope of access to Japanese postpositions.

본 연구는 도자 교육에 대한 연구동향을 파악하기 위하여 다양한 도자 교육 분야 저널 333종의 DB 를 제공하는 Art full text를 이용하였다. 1983년부터 2014년까지 도자 교육 분야의 주요 해외 저널에 게재된 논문의 2,711개 키워드를 수집한 후, KH Coder를 이용하여 연결 중심성, 매개 중심성, 위세 중심성등 중심성 분석을 수행하였다. 이를 통해 도자교육 분야 논문에 대한 구체적인 연구주제와 키워드 간의관계를 파악함으로써 학제적 연구와 융합을 위한 연구방향을 제시하고자 하였다. 또한 Qin(1999)의 다차원 척도화 결과와 사분면 키워드의 연구를 조합하여 도자교육의 연구동향을 살펴보았다. KH Coder와 다차원 척도화 사분면을 조합하여 다음과 같은 결론을 도출하였다. 첫째, 주류 연구는 위세 중심성이 높은 키워드와 1사분면 키워드의 연구를 진행하는 것이 연구가치가 높다고 할 수 있다. 둘째, 심도 연구는 연결 중심성이 높은 키워드와 2사분면 키워드의 연구를 진행하는 것이 연구가치가 높다고 할 수 있다. 셋째, 융합 연구는 매개 중심성이 높은 키워드와 3사분면 키워드의 연구를 진행하는 것이 연구가치가 높다고 할 수 있다. 넷째, 차세대 연구는 최근 1~2년 사이 새롭게 등장한 주류 키워드와 4사분면 키워드의 연구를 진행하는 것이 연구가치가 높다고 할 수 있다. 이와 같은 분석 결과는 도자 교육의 연구 동향을 처음으로 분석하였고, 도자 교육의 미래연구 방향을 제시하였다는 학문적 의의가 있다고 할 수 있다.

This study used the Art full text DB to provide a variety of apprenticeship 333 kinds of education journals to identify research trends in ceramic education. From 1983 to 2014 after gathering the keywords of the 2,711 papers published in international journals in the field of education ceramics, using KH Coder degree centrality, betweeness centrality, eigenvector centrality, etc. centrality analysis was performed. Thus, by knowing the relationship between the specific research themes of the papers of apprenticeship education and keywords, I tried to present the direction of research for interdisciplinary research and fusion. In addition, we discussed the research trend of ceramics education by combining the study of the results and the quadrant keyword of multidimensional scaling of Qin (1999). A combination of the KH Coder and multidimensional scaling quadrant, was derive the following conclusion. First, the mainstream of the research can be to proceed with the study of the eigenvector centrality keyword and 1 quadrant of the keyword is that there is a high value of research. Secondly, in the depth of research, it is possible to proceed with the study of the degree centrality keywords and 2 quadrant of the keyword is that there is a high value of research. Thirdly, the fusion research can be advanced research betweenness centrality keyword and 3 quadrants of the keyword and more valuable research. Fourth, the next generation of research can be advanced new study appeared in mainstream keywords and 4 quadrants of keywords in recent 1-2 years and a high value of research. The result of such analysis, first analyzes the research trends of ceramics education, it can be that there is a learning significance that presented a future research directions of ceramics education.

This study empirically examines the conceptual structure of negotiation in everyday language in Japan and South Korea through a text-mining analysis of Q&A websites. It assesses whether the academic trends discussed in Choi (2025) align with citizens' concerns, thereby uncovering the underlying cultural paradigms of both nations. The analysis reveals distinct patterns: Japanese discourse tends to prioritize "restoring order" and "self-defense" in negative contexts, such as disputes. In contrast, Korean discourse focuses on "acquiring economic interests" and "maximizing value," particularly in areas like salary and assets. Second, these differences reflect distinct cultural paradigms. Japanese negotiation is characterized by an "interdependent self-construal," where negotiation is seen as a strategic tool to achieve specific goals, separating the task from the emotional context. Third, network structures illustrate differing social frameworks. In South Korea, labor-related clusters demonstrate that negotiation is perceived as a strategic arena for asserting professional rights. In Japan, the focus is on the "self" in accordance with institutional norms (such as marriage and law enforcement), prioritizing compliance with these institutions over individual rights. This study shows that communication issues between Japan and South Korea arise not merely from linguistic differences but also from distinct cognitive frameworks governing language use. The findings highlight the need for future interpersonal communication education between the two countries to move beyond traditional language instruction. Instead, it should focus on developing the meta-cognitive awareness of each other’s socio-cultural cognitive structures.

This paper is a follow-up study of previous studies that analyzed the tourism use of domestic literature museums using text mining. Its purpose is to analyze the tourism use of literature museums in Japan to find ways to use them as local tourism facilities. First, through ‘Listly’ and ‘Python’, the search keyword ‘文学館 (Literary Museum)’ was entered on Japanese travel sites ‘Jaran(じゃらん)’ and ‘4travel’ to collect reviews of ‘Kuchi Komi (below review)’. Japan was divided into eight regions, and the review data collected accordingly were classified. After that, the co-occurrence network was created with KH Coder and LDA topic modeling was performed with Python. As a result, it has been confirmed that it is used as a local tourism resource in Japan through the exterior of the building of the literature museum, garden facilities, natural scenery, secondary contents, theme parks, and the restoration of birthplace. Through this paper, it was possible to grasp the tourism application patterns of Japanese literary museums.

This study aims to compare the cultural perceptions of the ‘negotiation’ between the Japanese and Korean academic communities. Using a quantitative text-mining approach, we analyze academic paper titles published from 2020 to 2025 in both languages. Through frequency analysis, co-occurrence network analysis, and correspondence analysis, we examine the structural framework of negotiation discourse and how it is conceptualized in each linguistic and cultural context. The results reveal that while both Japanese and Korean academic communities regard negotiation as a structured speech act, the organization of central themes and key terms varies significantly. Japanese titles tend to frame negotiation within formal and institutional contexts, such as labor relations and contracts. In contrast, Korean titles often reflect practical and personal contexts, integrating negotiation into everyday experiences. These findings suggest that perceptions of negotiation are deeply connected to each culture’s social values and interactional norms. The study emphasizes the need for communication education to address these cultural variations in the structural framework of negotiation discourse. The proposed multidimensional approach is expected to support the development of culturally responsive communication education programs.