• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2018년도 제30회 한글 및 한국어 정보처리 학술대회
• /
• pp.636-638
• /
• 2018

본 논문에서는 영어권에 비해 상대적으로 부족한 한국어 언어자원을 지속적으로 구축함으로써 한국어 문서로 구성된 시간정보 주석 말뭉치를 확보하고 이를 바탕으로 한국어 시간정보추출 시스템에 대한 연구를 수행한다. 말뭉치 구축 과정에서의 시간정보 주석 작업은 가이드라인을 숙지한 주석자들이 수작업으로 기록하고, 어떤 주석 결과에 대해 의견이 다른 경우에는 중재자가 주석자들과 함께 검토하며 합의점을 도출한다. 시간정보추출 시스템은 자연어 문장에 대한 형태소 분석결과를 이용하여 시간표현(TIMEX3), 시간관계와 연관된 사건(EVENT), 시간표현 및 사건들 간의 시간관계(TLINK)를 추출하는 단계로 이루어진다. 추출된 한국어 시간정보는 문서 내 공통된 개체에 대한 공간정보와 결합함으로써 시공간정보가 모두 반영된 SPOTL을 생성한다. 추후 실험을 통하여 제안시스템의 구체적인 시간정보추출 성능을 파악할 것이다.

• 영어어문교육
• /
• 제5호
• /
• pp.79-101
• /
• 1999

With the advent of the 6th and the 7th national educational curricula in Korea. English language teaching in communicative perspective has been highly recommended and widely practised in that context. The aim of new approach is to enhance the students' general communication abilities in English. However, English teachers still find it very hard to improve the students' communicative competence in English since English remains to be taught as a school subject. In so far as English is taught as a school subject, students' attention is paid to the formal elements of English and the increase in communicative competence in English is hardly expected Only when the students' attention is paid to the content, their communicative competence is expected to increase. The best way to shift the students' attention from formal elements to content is to teach other school subjects in English, that is, English immersion education. To introduce immersion education to Korea, the two most successful examples of Singapore and Hong Kong are reviewed in terms of language policies and general practices in their primary and secondary schools respectively. To implement the program into Korea, extensive research on it is expected henceforth.

• 한국언어정보학회:학술대회논문집
• /
• 한국언어정보학회 2007년도 정기학술대회
• /
• pp.285-292
• /
• 2007

The parallel corpus is an important resource in the research field of data-driven natural language processing, but there are only a few parallel corpora publicly available nowadays, mostly due to the high labor force needed to construct this kind of resource. A novel strategy is brought out to automatically fetch parallel text from the web in this paper, which may help to solve the problem of the lack of parallel corpora with high quality. The system we develop first downloads the web pages from certain hosts. Then candidate parallel page pairs are prepared from the page set based on the outer features of the web pages. The candidate page pairs are evaluated in the last step in which the sentences in the candidate web page pairs are extracted and aligned first, and then the similarity of the two web pages is evaluate based on the similarities of the aligned sentences. The experiments towards a multilingual web site show the satisfactory performance of the system.

• 영어어문교육
• /
• 제18권3호
• /
• pp.149-174
• /
• 2012

This research aims at exploring what Korean teachers of English need for their effective professional development, particularly from the perspectives of peer trainers who are school teachers as well as teacher trainers. The research is based on an assumption that despite the critical roles of peer trainers in the changing process of English education, very little attention has been paid to listening to their voices. Having an awareness of the meaningfulness of their voices in teacher education, the researcher collected and analyzed the data, employing a questionnaire responded by 32 peer trainers and in-depth group interview with 6 peer trainers who responded the questionnaire. The research findings suggest that the majority of peer trainers believe that the priority should be put on teaching practice rather than language competence for Korean English language teacher development. It has been found that they believe that teacher learning within the workplace with some support from outside experts is the most effective way of making classroom change possible and keeping teachers' professional development sustainable. Finally, the research suggests school-based professional development as an alternative strategy to satisfy their needs on effective and sustainable professional development.

• 영어어문교육
• /
• 제8권1호
• /
• pp.199-221
• /
• 2002

Multimedia-Assisted Language Learning(MALL) has recently been studied by many researchers. It has been reported that MALL is very effective in encouraging students' desire for learning, promoting their self-directed learning, and improving their communicative competence. Also, it has been evident that it depends on the quality of multimedia courseware whether MALL will be successful or not. However, many researchers have pointed out that most of multimedia coursewares coming into the market have little to do with our curriculum and they are not suitable for the use in the regular classroom. More multimedia coursewares that reflect our curriculum need to be developed. This paper first tries to explore the cognitive, constructivist, and psychological theories supportive of the development of multimedia courseware and then presents the overall procedure for designing and developing a multimedia courseware pursuant to the 7th English curriculum in the middle school. The multimedia courseware developed through this research is expected to enhance middle school students' communicative language skills in English and promote the development of multimedia coursewares of high quality.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2023년도 제35회 한글 및 한국어 정보처리 학술대회
• /
• pp.131-136
• /
• 2023

본 논문에서는 기존의 요약 태스크에서 주로 사용하는 인코더-디코더 모델과 디코더 기반의 언어 모델의 성능을 비교한다. 요약 태스크를 평가하는 주요한 평가 지표인 ROUGE 점수의 경우, 정답 요약문과 모델이 생성한 요약문 간의 겹치는 단어를 기준으로 평가한다. 따라서, 추상적인 요약문을 생성하는 언어 모델의 경우 인코더-디코더 모델에 비해 낮은 ROUGE 점수가 측정되는 경향이 있다. 또한, 최근 연구에서 정답 요약문 자체의 낮은 품질에 대한 문제가 되었고, 이는 곧 ROUGE 점수로 모델이 생성하는 요약문을 평가하는 것에 대한 신뢰도 저하로 이어진다. 따라서, 본 논문에서는 언어 모델의 요약 성능을 보다 다양한 관점에서 평가하여 언어 모델이 기존의 인코더-디코더 모델보다 좋은 요약문을 생성한다는 것을 보인다.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2023년도 제35회 한글 및 한국어 정보처리 학술대회
• /
• pp.312-318
• /
• 2023

본 연구는 다양한 대규모 언어 모델들의 한국어/영어 공감 대화 생성에서 성능을 실험적으로 비교 분석하는 것과 개인의 성향과 공감 사이에서의 상관 관계를 실험적으로 분석하는 것을 목표로 한다. 이를 위해, 한국어 공감 대화 데이터셋인 KorEmpatheticDialogues 를 구축하였고, personality-aware prompting 방법을 제안한다. 실험을 통해, 총 18개의 언어 모델들 간의 공감 대화 생성 성능을 비교 분석하였고, 개인의 성향에 맞춤형 제공하는 공감이 더 상호작용을 이끌어낼 수 있다는 점을 보여준다. 코드와 데이터셋은 게재가 허용되면 공개할 예정이다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2008년도 추계학술발표대회
• /
• pp.168-170
• /
• 2008

한글 상호(商號) 로마자 변환기는 한글로 만들어진 상호를 로마자로 자동 변환하는 시스템이다. 이 변환기는 기사용 로마자 상호명과 업종명, 그리고 표준 한글 로마자 변환 규칙에 의해 생성한 로마자를 조합하여 로마자 상호를 생성한다. 이때, 조합을 위한 알고리즘이 필요한데, 기존에 비슷한 용도에 사용되었던 stack 알고리즘을 적용할 경우 비효율적이다. 본 논문은 이를 대체할 새 알고리즘을 제안한다. 새 알고리즘은 기존 stack 알고리즘을 사용할 때에 비해 복잡도를 O(b^d)에서 O(b^*d)로 줄여 성능을 높인다.

• Asia pacific journal of information systems
• /
• 제30권4호
• /
• pp.832-846
• /
• 2020

Service organizations increasingly adopt data-based intelligent engines called chatbots in support of the interaction between customers and the companies. Two different types of chatbots have been suggested and introduced by companies leading the adoption of this emerging technology: rule-based chatbots and natural language processing-based chatbots. While the differences between these two types of technologies look relatively clear, the organizational and practical impacts of the differences have not been systematically explored. This study performed an experiment to compare the use of the two different types of chatbots used in practice by two comparable organizations. These two types of actual chatbots were used by Korean on-line shopping malls with similar business models (mobile shopping), length of history, size and reputation. The comparison was made based on such dimensions as usability, searchability, reliability and attractiveness. Contraty to conventional expectation that the superiority in technology will produce superior usability, the results show mixed superiority. The discussion on the reasons is presented.

• Korean Journal of Health Education and Promotion
• /
• 제2권1호
• /
• pp.1-19
• /
• 2000

With rapid economic development, the emphasis of the public health movement in Korea has shifted towards addressing the burden of chronic disease. With this shift in direction comes a greater focus on health behaviour and the need for planning models to assist in lifestyle modification programs. The Health Belief Model (HBM), which originated in the US, has generated more research than any other theoretical approach to describe and predict the health behaviour of individuals. In recent years it has been applied in many different cultures and modifications have been suggested to accommodate different cultures. Given the centrality of language and culture, any attempts to use models of health behaviour developed in a different culture, must be studied and tested for local applicability. The paper reviews the applicability and suitability of the HBM in Korea, in the context of the Korean language and culture. The HBM has been used in Korea for almost three decades. The predictability of the HBM has varied in Korean studies as in other cultures. Overall, this literature review indicates that the HBM has been found applicable in predicting health and illness behaviours by Korean people. However if the HBM is used in a Korean context, the acquisition of health knowledge is an important consideration. Most new knowledge in the health sciences is originally published in English and less frequently in another foreign language. Most health knowledge in Korea is acquired through the media or from health professionals and its acquisition often involves translation from the original. The selection of articles for translation and the accuracy of translation into language acceptable in the Korean culture become important determinants of health knowledge. As such translation becomes an important part of the context of the HBM. In this paper modifications to the HBM are suggested to accommodate the issues of language and knowledge in Korea.