• International Journal of Contents
• /
• 제13권3호
• /
• pp.32-37
• /
• 2017

The development of Information and Communications Technology (ICT) has changed the manner in which English teachers are taught as instructors, as well as English language learners. Online-related technology, a product of development of ICT technology, is used for various purposes such as training teachers and enhancing professionalism of current teachers to practice more efficient training. The purpose of this study is to extract common domains and sub-domains related to development of cyber-type in-service training (INSET) through domestic and international case analyses and to develop content areas of INSET using cyberspace or efficient online INSET programs. To accomplish the purpose of this study, domestic and foreign cases were analyzed in relation to direction of development of online English teacher training programs.

• 영어어문교육
• /
• 제3호
• /
• pp.71-93
• /
• 1997

Teachers' beliefs about teaching English are reflected in their practices in the classroom. They influence on the students' attitude to English learning. Any teacher training program expects the trainees to change or modify their existing beliefs and attitude through the new ideas and information introduced by the program toward a desired direction. The present study describes a teacher training program for elementary school English teachers and compares the beliefs of the teachers about teaching English before and after the training. The subjects are the elementary school English teachers around Chungnam area who get a special training of 120 hours during January 1997. The investigation of the subjects' beliefs on English teaching is conducted through examining two journals of each subject before and after the training. The journals show the teachers' inner flow of thought, so teacher trainers are expected to get insight on their general instructional considerations and have implications on the future teacher training program through examining these journals. In addition, the journal writing itself gives the teachers opportunity to reflect their practice and rethink about their beliefs, and develop themselves as professional English teachers.

• 영어어문교육
• /
• 제15권2호
• /
• pp.151-168
• /
• 2009

The present study investigates the impact of trained peer feedback on the quantity and quality of revisions made by EFL students at a low-intermediate level. Peer review training was carried out in experimental group through four in-class training sessions and four peer dyad-instructor conferences after class. Students' $1^{st}$ drafts with written peer feedback and revised drafts prior to and post training were collected and analyzed. Results reveal that after training the students produced more revisions in response to their peer comments (96% of total revisions) and those revisions were counted as enhanced in quality (93% of peer-triggered revisions). In contrast, the results of paired t-test within control group indicate that there was no significant difference between two data collected from week 3 and week 16 (t = -.57, df =19, p = .577 at p < .05). The findings suggest that training as an ongoing process of teacher intervention contributes to effectiveness of the peer feedback activity. The study provides pedagogical implications for how to structure and implement peer review training for the sake of its direct strength in an EFL writing class.

• 대한치과기공학회지
• /
• 제34권3호
• /
• pp.303-314
• /
• 2012

Purpose: This study aimed to find overseas workplace and improve global competence through the preference survey on overseas employment by dental technology students. Methods: The survey sample consisted of 250 randomly selected dental technology students. Survey was conducted from March 1 to May 1 in 2012. Total of 245(98.0%) replies and analyzed 236 questionnaires excluding 9 incomplete questionnaires. The questionnaires used in this study consisted of 7 items for general information, overseas employment characteristics of 10 items, 7 items for overseas employment activation plane and job competency development of 7 items. Collected data were analyzed using SPSS(Statistical Package for Social Sciences) Win 19.0 statistics program. Results: Regarding general characteristics of the subjects, there were 131 third graders(55.5%), 63 first graders(26.7%) and 42 second graders(17.8%) among 130 males(55.1%) and 106 females(44.9%). 221(93.6%) of the subjects had no experience in language training. Students who had clinical training for 1-5 months were 123(52.1%), and 24(10.2%) students had more than six months. 89(37.7%) of the subjects had no clinical training. 155(65.7%) of the subjects hope to work with korean owner, and 81(34.3%) chose foreign owner. Favored working countries were Australia(41.5%), the United States(29.2%), Canada(18.2%) and other(11.0%). The field of dental ceramic was indicated to be the highest proportion of 104(44.1%). Period of training were 3 hours(40.3%) and 6 hours(35.2%). The most important training were language-centered education(54.2%), Job-oriented education(24.2%), local culture education(16.1%), other(3.0%) and Leadership Training(2.5%). The subjects chose overseas worker(44.9%), working-level practitioner (28.8%), successfully employed dental technology graduate(19.5%a) and professor(3.4%) as an instructor. The subjects get education and training information from professor(40.3%), other(28.0%), senior(14.4%), job site(8.9%) and acquaintance(8.5%). A credit exchange(2.46 points), a joint degree program(2.46 points), and a foreign professor(2.33 points) were needed to activate the overseas employment. A kind of dental prosthesis(3.58 points), carving tooth morphology(3.38 points), and majors of dental technology(3.30 points) were indicated to develop job competency for overseas employment. Age, year, clinical training experience and company owner were statistically meaningful data among the general characteristics affecting job competency development. Conclusion: The college needs to offer variety programs such as foreign language-centered education and a local job competency development program to graduates to be connected with international workplace and employment.

• 지능정보연구
• /
• 제23권2호
• /
• pp.71-88
• /
• 2017

언어모델은 순차적으로 입력된 자료를 바탕으로 다음에 나올 단어나 문자를 예측하는 모델로 언어처리나 음성인식 분야에 활용된다. 최근 딥러닝 알고리즘이 발전되면서 입력 개체 간의 의존성을 효과적으로 반영할 수 있는 순환신경망 모델과 이를 발전시킨 Long short-term memory(LSTM) 모델이 언어모델에 사용되고 있다. 이러한 모형에 자료를 입력하기 위해서는 문장을 단어 혹은 형태소로 분해하는 과정을 거친 후 단어 레벨 혹은 형태소 레벨의 모형을 사용하는 것이 일반적이다. 하지만 이러한 모형은 텍스트가 포함하는 단어나 형태소의 수가 일반적으로 매우 많기 때문에 사전 크기가 커지게 되고 이에 따라 모형의 복잡도가 증가하는 문제가 있고 사전에 포함된 어휘 외에는 생성이 불가능하다는 등의 단점이 있다. 특히 한국어와 같이 형태소 활용이 다양한 언어의 경우 형태소 분석기를 통한 분해과정에서 오류가 더해질 수 있다. 이를 보완하기 위해 본 논문에서는 문장을 자음과 모음으로 이루어진 음소 단위로 분해한 뒤 입력 데이터로 사용하는 음소 레벨의 LSTM 언어모델을 제안한다. 본 논문에서는 LSTM layer를 3개 또는 4개 포함하는 모형을 사용한다. 모형의 최적화를 위해 Stochastic Gradient 알고리즘과 이를 개선시킨 다양한 알고리즘을 사용하고 그 성능을 비교한다. 구약성경 텍스트를 사용하여 실험을 진행하였고 모든 실험은 Theano를 기반으로 하는 Keras 패키지를 사용하여 수행되었다. 모형의 정량적 비교를 위해 validation loss와 test set에 대한 perplexity를 계산하였다. 그 결과 Stochastic Gradient 알고리즘이 상대적으로 큰 validation loss와 perplexity를 나타냈고 나머지 최적화 알고리즘들은 유사한 값들을 보이며 비슷한 수준의 모형 복잡도를 나타냈다. Layer 4개인 모형이 3개인 모형에 비해 학습시간이 평균적으로 69% 정도 길게 소요되었으나 정량지표는 크게 개선되지 않거나 특정 조건에서는 오히려 악화되는 것으로 나타났다. 하지만 layer 4개를 사용한 모형이 3개를 사용한 모형에 비해 완성도가 높은 문장을 생성했다. 본 논문에서 고려한 어떤 시뮬레이션 조건에서도 한글에서 사용되지 않는 문자조합이 생성되지 않았고 명사와 조사의 조합이나 동사의 활용, 주어 동사의 결합 면에서 상당히 완성도 높은 문장이 발생되었다. 본 연구결과는 현재 대두되고 있는 인공지능 시스템의 기초가 되는 언어처리나 음성인식 분야에서 한국어 처리를 위해 다양하게 활용될 수 있을 것으로 기대된다.

• 한국정보전자통신기술학회논문지
• /
• 제16권4호
• /
• pp.193-202
• /
• 2023

최근 대규모 데이터 세트로 학습된 거대 언어 모델들의 뛰어난 성능이 공개되면서 큰 화제가 되고 있다. 하지만 거대 언어 모델을 학습하고 활용하기 위해서는 초대용량의 컴퓨팅 및 메모리 자원이 필요하므로, 대부분의 연구는 빅테크 기업들을 중심으로 폐쇄적인 환경에서 진행되고 있었다. 하지만, Meta의 거대 언어 모델 LLaMA가 공개되면서 거대 언어 모델 연구들은 기존의 폐쇄적인 환경에서 벗어나 오픈 소스화되었고, 관련 생태계가 급격히 확장되어 가고 있다. 이러한 배경하에 사전 학습된 거대 언어 모델을 추가 학습시켜 특정 작업에 특화되거나 가벼우면서도 성능이 뛰어난 모델들이 활발히 공유되고 있다. 한편, 사전 학습된 거대 언어 모델의 학습데이터는 영어가 큰 비중을 차지하기 때문에 한국어의 성능이 비교적 떨어지며, 이러한 한계를 극복하기 위해 한국어 데이터로 추가 학습을 시키는 한국어 특화 언어 모델 연구들이 이루어지고 있다. 본 논문에서는 오픈 소스 기반의 거대 언어 모델의 생태계 동향을 파악하고 영어 및 한국어 특화 거대 언어 모델에 관한 연구를 소개하며, 거대 언어 모델의 활용 방안과 한계점을 파악한다.

• 전자통신동향분석
• /
• 제38권6호
• /
• pp.1-11
• /
• 2023

Large language models seem promising for handling reasoning problems, but their underlying solving mechanisms remain unclear. Large language models will establish a new paradigm in artificial intelligence and the society as a whole. However, a major challenge of large language models is the massive resources required for training and operation. To address this issue, researchers are actively exploring compact large language models that retain the capabilities of large language models while notably reducing the model size. These research efforts are mainly focused on improving pretraining, instruction tuning, and alignment. On the other hand, chain-of-thought prompting is a technique aimed at enhancing the reasoning ability of large language models. It provides an answer through a series of intermediate reasoning steps when given a problem. By guiding the model through a multistep problem-solving process, chain-of-thought prompting may improve the model reasoning skills. Mathematical reasoning, which is a fundamental aspect of human intelligence, has played a crucial role in advancing large language models toward human-level performance. As a result, mathematical reasoning is being widely explored in the context of large language models. This type of research extends to various domains such as geometry problem solving, tabular mathematical reasoning, visual question answering, and other areas.

• 한국컴퓨터정보학회논문지
• /
• 제25권4호
• /
• pp.29-35
• /
• 2020

본 논문에서는 새로운 구조의 개체명 인식 DeNERT 모델을 제안한다. 최근 자연어처리 분야는 방대한 양의 말뭉치로 사전 학습된 언어 표현 모델을 활용하는 연구가 활발하다. 특히 자연어처리 분야 중 하나인 개체명인식은 대부분 지도학습 방식을 사용하는데, 충분히 많은 양의 학습 데이터 세트와 학습 연산량이 필요하다는 단점이 있다. 강화학습은 초기 데이터 없이 시행착오 경험을 통해 학습하는 방식으로 다른 기계학습 방법론보다 조금 더 사람이 학습하는 과정에 가까운 알고리즘으로 아직 자연어처리 분야에는 많이 적용되지 않은 분야이다. 아타리 게임이나 알파고 등 시뮬레이션 가능한 게임 환경에서 많이 사용된다. BERT는 대량의 말뭉치와 연산량으로 학습된 구글에서 개발한 범용 언어 모델이다. 최근 자연어 처리 연구 분야에서 높은 성능을 보이고 있는 언어 모델이며 많은 자연어처리 하위분야에서도 높은 정확도를 나타낸다. 본 논문에서는 이러한 DQN, BERT 두가지 딥러닝 모델을 이용한 새로운 구조의 개체명 인식 DeNERT 모델을 제안한다. 제안하는 모델은 범용 언어 모델의 장점인 언어 표현력을 기반으로 강화학습 모델의 학습 환경을 만드는 방법으로 학습된다. 이러한 방식으로 학습된 DeNERT 모델은 적은 양의 학습 데이터세트로 더욱 빠른 추론시간과 높은 성능을 갖는 모델이다. 마지막으로 제안하는 모델의 개체명 인식 성능평가를 위해 실험을 통해서 검증한다.

• The Journal of the Acoustical Society of Korea
• /
• 제23권2E호
• /
• pp.51-55
• /
• 2004

In this paper, we propose a style-specific language model adaptation scheme using n-gram based tf*idf similarity for Korean spontaneous speech recognition. Korean spontaneous speech shows especially different style-specific characteristics such as filled pauses, word omission, and contraction, which are related to function words and depend on preceding or following words. To reflect these style-specific characteristics and overcome insufficient data for training language model, we estimate in-domain dependent n-gram model by relevance weighting of out-of-domain text data according to their n-. gram based tf*idf similarity, in which in-domain language model include disfluency model. Recognition results show that n-gram based tf*idf similarity weighting effectively reflects style difference.

• 영어어문교육
• /
• 제15권2호
• /
• pp.71-92
• /
• 2009

Europass was established in 2005 by the European Parliament and the Council of Europe as a single framework for language qualifications and competences, helping citizens to gain accreditation throughout the European Community. In addition, the 1996 Common European Framework of Reference for Languages: Learning, Teaching, Assessment (CEFR) provides a common basis for language syllabi, curriculum guidelines, examination, and textbooks in Europe. This framework describes the required knowledge and skills, the cultural context, and the levels of proficiency that learners should achieve. In combination, Europass and the CEFR provide employers and educational institutes with internationally recognized standards. This paper proposes that current trends such as globalization and international mobility require a similar approach to accreditation in Asia. As jobs and workers become independent of national boundaries and restrictions, it becomes necessary to educate students as multilingual world citizens, using standards that are accepted around the world. It is suggested, therefore, that assessment models such as Europass and the CEFR, along with successful language teaching models in Europe and Canada, present opportunities of adaptation for the Korean education system. Finally, rigorous teacher training to internationally recognized levels is recommended, if Korea is to produce a workforce of highly-skilled, plurilingual world citizens.