http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : 이해미 (검색결과 99 건)
- 무료
- 기관 내 무료
- 유료
-
-
-
-
-
-
이해미,장은재 한국식생활문화학회 2001 한국식생활문화학회지 Vol.16 No.5
This study was conducted to investigate the effects of corn peptide(CP) on lipid metabolism. Sprague-Dawley(S.D) male rats were assigned to three dietary groups {control diet(CD), high fat diet(HFD) & high fat com peptide diet(FCD)}l and fed 4 weeks to examine the effects of CP. There were no significantly different in cholesterol concentrations in the liver among the groups. However, triglyceride(TG) concentrations of the FCD & CD significantly lower than the HFD. Fecal excretion of neutral steroids & bile acids of the FCD significantly higher than the CD & HFD. Serum total cholesterol, TG & LDL-cholesterol concentrations of the FCD & CD significantly lower than the HFD. These results suggest the improvement of lipid composition in serum by CP might be inhibit of lipid absorption in intestine & increment of neutral steroids & bile acids excretion in feces.
-
일본 신문 사설 헤드라인과 본문의 응집성 분석 키워드・클레임 구조와 응집성 메커니즘을 중심으로
이해미 동국대학교 일본학연구소 2025 일본학 Vol.66 No.-
본고는 일본 주요 신문 3사의 사설 4,116편을 대상으로 헤드라인과 본문 간 응집성을 분석하였다. 기존 연구가 소규모 텍스트 분석에 그쳤던 한계를 극복하기 위해 대규모 코퍼스에 문자・단어・의미 단위의 3단계 매칭 분석을 적용하여 응집성 패턴을 체계적으로 규명하였다. 분석 결과 일본 신문 사설은 키워드와 클레임으로 구성된 이원적 헤드라인 구조를 일관되게 유지하는 것으로 나타났다. 특히 주목할 만한 발견은 응집성 구현 방식이 담화 진행에 따라 분화된다는 점이다. 문자 및 단어 단위 매칭은 도입부와 초반부에 집중되어 주제를 명확히 제시하는 반면, 의미 단위 매칭은 중반부에 집중되어 개념적 논의를 심화하는 역할을 담당한다. 이러한 단계별 응집성 전략의 분화는 응집성이 정적인 텍스트 결속이 아닌 담화 기능에 따라 분화되는 동적 메커니즘임을 보여준다. 본고는 기존 응집성 이론을 확장하는 관점을 제시하며, 텍스트마이닝을 활용한 담화분석의 새로운 방법론을 구축하였다. This study analyzes cohesion between headlines and body text in 4,116 editorials from three major Japanese newspapers. To overcome the limitations of existing research that focused on small-scale text analysis, this study applied a three-level matching analysis (character, word, and semantic levels) to a large-scale corpus to systematically identify cohesion patterns. The analysis reveals that Japanese newspaper editorials consistently maintain a binary headline structure composed of keywords and claims. A particularly notable finding is that cohesion implementation strategies differentiate according to discourse progression. Character and word-level matching concentrate in the introduction and early sections to clearly present topics, while semantic-level matching concentrates in the middle sections to deepen conceptual discussions. This differentiation of cohesion strategies by discourse stages demonstrates that cohesion functions as a dynamic mechanism that differentiates according to discourse functions, rather than static textual binding. This study presents a new perspective that extends existing cohesion theory and establishes an innovative methodology for discourse analysis utilizing text mining techniques.
-
-
일본어 전공 대학생의 생성형 AI 활용 실태와 인식 변화 분석- 2023-2025년 변화 양상을 중심으로 -
이해미 한국일본어교육학회 2026 일본어교육 Vol.- No.115
This study examines the usage patterns and perceptions of generative AI among Japanese language major undergraduates at a Korean university, focusing on the 2025 survey (n=142) while drawing on the 2023 survey (n=152) as a reference point for tracking change. While the first survey captured early-stage adoption one year after ChatGPT's release—with only 18.4% of students using AI regularly—the follow-up survey revealed a dramatic shift: 100% ChatGPT awareness, 77.5% usage rates, 64.7% biweekly use, and a diversified translation tool ecosystem incorporating AI alongside Papago. Key findings point to a central paradox: learners report high satisfaction (95.1% for time efficiency) yet harbor deep concerns over cognitive dependency (68.3%) and creativity decline (63.4%). AI use is heavily concentrated in translation (47.9%) and writing (42.3%)—core cognitive processes in foreign language learning—while listening and speaking remain largely untouched. Institutionally, 57.0% of students are unaware of university AI guidelines and 63.4% have received no formal training, despite having developed their own ethical norms around agency and transparency. These findings suggest three directions for pedagogical response: process-centered task design that foregrounds prompt construction and critical error detection; Japanese-specific AI literacy education addressing model limitations in honorifics and context-dependent expression; and policy frameworks that build on students' emergent norms rather than imposing external standards. This study offers baseline data for foreign language education navigating rapid AI diffusion, where broad adoption has outpaced both institutional readiness and learner self-awareness.
-
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
