• 한국IT서비스학회지
• /
• 제19권5호
• /
• pp.83-91
• /
• 2020

Language Models such as BERT has been an important factor of deep learning-based natural language processing. Pre-training the transformer-based language models would be computationally expensive since they are consist of deep and broad architecture and layers using an attention mechanism and also require huge amount of data to train. Hence, it became mandatory to do fine-tuning large pre-trained language models which are trained by Google or some companies can afford the resources and cost. There are various techniques for fine tuning the language models and this paper examines three techniques, which are data augmentation, tuning the hyper paramters and partly re-constructing the neural networks. For data augmentation, we use no-answer augmentation and back-translation method. Also, some useful combinations of hyper parameters are observed by conducting a number of experiments. Finally, we have GRU, LSTM networks to boost our model performance with adding those networks to BERT pre-trained model. We do fine-tuning the pre-trained korean-based language model through the methods mentioned above and push the F1 score from baseline up to 89.66. Moreover, some failure attempts give us important lessons and tell us the further direction in a good way.

• 아동학회지
• /
• 제29권2호
• /
• pp.155-176
• /
• 2008

This study explored variables related to Korean language education for preschool children with multicultural family backgrounds. Participants were 21 Korean language teachers and 14 women who immigrated from China, Japan, Mongolia, Philippines, and Vietnam to marry Korean men. They were mothers of children 2 to 7 years of age and had lived in Korea an average of five years. Mean age of mothers was 37(range of 30 to 43). Half had college and none had less then middle school education. They were interviewed with a series of semi-structured questionnaires. The children were reported to have a low level of vocabulary and articulation because their mothers could not provide fruitful oral language experiences. Supporting systems including family literacy were discussed.

• Journal of Information Processing Systems
• /
• 제4권4호
• /
• pp.121-132
• /
• 2008

VLSI chips have been tested using various automatic test equipment (ATE). Although each ATE has a similar structure, the language for ATE is proprietary and it is not easy to convert a test program for use among different ATE vendors. To address this difficulty we propose a tester structure expression language, a tester language with a novel format. The developed language is called the general tester language (GTL). Developing an interpreter for each tester, the GTL program can be directly applied to the ATE without conversion. It is also possible to select a cost-effective ATE from the test program, because the program expresses the required ATE resources, such as pin counts, measurement accuracy, and memory capacity. We describe the prototype environment for the GTL and the tester selection tool. The software size of the prototype is approximately 27,800 steps and 15 manmonths were required. Using the tester selection tool, the number of man-hours required in order to select an ATE could be reduced to 1/10. A GTL program was successfully executed on actual ATE.

• Journal of Information Processing Systems
• /
• 제18권4호
• /
• pp.535-548
• /
• 2022

Using a rich resource language to classify sentiments in a language with few resources is a popular subject of research in natural language processing. Burmese is a low-resource language. In light of the scarcity of labeled training data for sentiment classification in Burmese, in this study, we propose a method of transfer learning for sentiment analysis of a language that uses the feature transfer technique on sentiments in English. This method generates a cross-language word-embedding representation of Burmese vocabulary to map Burmese text to the semantic space of English text. A model to classify sentiments in English is then pre-trained using a convolutional neural network and an attention mechanism, where the network shares the model for sentiment analysis of English. The parameters of the network layer are used to learn the cross-language features of the sentiments, which are then transferred to the model to classify sentiments in Burmese. Finally, the model was tuned using the labeled Burmese data. The results of the experiments show that the proposed method can significantly improve the classification of sentiments in Burmese compared to a model trained using only a Burmese corpus.

• International Journal of Computer Science & Network Security
• /
• 제22권11호
• /
• pp.63-70
• /
• 2022

At the present stage, the main directions of the professional position of a specialist in the implementation of English-language Education are to improve and spread the practice of learning languages throughout a person's life by involving information, communication and digital technologies in the educational process. Computerization of the educational process in Higher Education Institutions is considered as one of the first and most promising areas for improving the quality of education in Higher Education Institutions. The necessity of ensuring timely training and retraining of specialists of various profiles (in particular teachers) on the effective use of domestic and foreign electronic resources with the help of modern information technologies for the implementation of the professional position of a future specialist in a foreign-language environment is noted. The main goal of teaching a foreign language (the formation of students' communicative competence, which means mastering the language as a means of intercultural communication) is defined. The types of speech activity that cover the content of teaching a foreign language are highlighted. The main types of assessment in a foreign language are shown - current (non-classroom), thematic, semester, annual assessment and final state certification. The task of the teacher is drawn, which is to create conditions for practical language acquisition for each student, to choose such teaching methods by means of information technologies that would allow each student to show their activity, their creativity; to activate the cognitive activity of the student in the process of learning a foreign language.

• 한국비블리아학회지
• /
• 제34권3호
• /
• pp.227-253
• /
• 2023

이 연구는 디지털 혁명으로 인해 변화한 정보 환경에서 리터러시 교육을 위한 정보자원으로 밈을 개념화하기 위한 시론적 연구이다. 이 연구의 목적은 정보자원으로서 밈의 활용을 촉진하기 위해 밈의 맥락과 실재를 규명하는 데에 있다. 이를 연구문제로 환원시키면 다음과 같다. 첫째, '밈'은 어떠한 주제들과 함께 연구되는가? 둘째, 어떠한 것들이 '밈'으로 포착되어 연구되는가? 연구문제를 소명하기 위해 국내 밈 연구 145건을 대상으로 빈도분석과 동시출현빈도 네트워크 분석을 실시하였고, 이 중 73건을 대상으로 밈 사례 275개를 추출하여 내용 분석을 실시하였다. 그 결과는 다음과 같다. 첫째, 밈은 인문학, 사회과학, 복합학, 예술체육학 분야에서 주로 연구되고 있었다. 또한 연구 초기에는 Dawkins의 밈 개념을 토대로 한 이론 연구(2012년 전후), 한류 콘텐츠 확산을 설명하기 위한 밈 개념 도입 연구(2015년 전후), 문화사회학의 주요 연구 주제로서 밈 개념의 독자적 연구(2019년 전후)가 이루어졌다. 둘째, 밈은 언어적 특성을 중점적으로 가지고 있었다. 언어 밈(L-meme)(102건, 37%)과 언어·시각 밈(LV-meme)(23건, 8%), 언어·시·청각 밈(LVM-meme)(21건, 8%) 등 언어 양식에 기반한 밈이 다수였고, 동시출현빈도 네트워크의 빈도·연결중심성·매개중심성 상위 노드에도 language meme(언어 밈) 키워드가 등장했다. 즉, 밈은 언어적 특성을 토대로 한 문화사회학의 고유한 정보 현상 개념으로 확장되고 있다. 리터러시 교육을 위한 정보자원으로서뿐만이 아니라 정보 리터러시의 관점에서 밈 리터러시를 개념화할 필요가 있다.

• 한국컴퓨터정보학회논문지
• /
• 제29권3호
• /
• pp.43-54
• /
• 2024

최근 언어모델을 활용하기 위한 연구가 활발히 이루어지며, 큰 규모의 언어모델이 다양한 과제에서 혁신적인 성과를 달성하고 있다. 하지만 실제 현장은 거대 언어모델 활용에 필요한 자원과 비용이 한정적이라는 한계를 접하면서, 최근에는 주어진 자원 내에서 모델을 효과적으로 활용할 수 있는 방법에 주목하고 있다. 대표적으로 학습 데이터를 난이도에 따라 구분한 뒤 순차적으로 학습하는 방법론인 커리큘럼 러닝이 주목받고 있지만, 난이도를 측정하는 방법이 복잡하거나 범용적이지 않다는 한계를 지닌다. 따라서, 본 연구에서는 신뢰할 수 있는 사전 정보를 통해 데이터의 학습 난이도를 측정하고, 이를 다양한 과제에 쉽게 활용할 수 있는 데이터 이질성 기반 커리큘럼 러닝 방법론을 제안한다. 제안방법론의 성능 평가를 위해 국가 R&D 과제 전문 문서 중 정보통신 분야 전문 문서 5,000건, 보건의료전문 문서 데이터 4,917건을 적용하여 실험을 수행한 결과, 제안 방법론이 LoRA 미세조정과 전체 미세조정 모두에서 전통적인 미세조정에 비해 분류 정확도 측면에서 우수한 성능을 나타냄을 확인했다.

• 디지털콘텐츠학회 논문지
• /
• 제7권3호
• /
• pp.175-182
• /
• 2006

The Internet is now of the fastest growing areas of telecommunications and of Computer Assisted Language Learning. It is rapidly becoming more integrated into society and accessible to people form around the world. A number of educators believe there is potential for language teaching and learning opportunities through the Internet, and have already developed uses and resources for this purpose. The range of what is available is growing continually. The purpose of this study is to research CMC via the Internet and other long-distance networks, to investigate the analyse best and worst things about studying English on the internet and to suggest some findings from the comparison between internet and classroom learning by means of questionnaire.

• Journal of information and communication convergence engineering
• /
• 제18권4호
• /
• pp.207-215
• /
• 2020

Part of speech (POS) tagging is an indispensable part of major NLP models. Its progress can be perceived on number of languages around the globe especially with respect to European languages. But considering Indian Languages, it has not got a major breakthrough due lack of supporting tools and resources. Particularly for Odia language it has not marked its dominancy yet. With a motive to make the language Odia fit into different NLP operations, this paper makes an attempt to develop a POS tagger for the said language on a HMM (Hidden Markov Model) platform. The tagger judiciously considers bigram HMM with dynamic Viterbi algorithm to give an output annotated text with maximum accuracy. The model is experimented on a corpus belonging to tourism domain accounting to a size of approximately 0.2 million tokens. With the proportion of training and testing as 3:1, the proposed model exhibits satisfactory result irrespective of limited training size.

• 한국어교육
• /
• 제29권2호
• /
• pp.73-100
• /
• 2018

The purpose of this study is to investigate learner-learner dialogue during speaking tasks. In the Korean language classroom, conversation between learners is an important activity as speaking practice. However, learner dialogue is also a tool to enable learners to collaboratively conduct various cognitive activities in the classroom. In previous research, it was unfolded that through learner-learner dialogue, learners can solve second-language related problems and set a goal to carry out tasks. Therefore, this study analyzed learner-learner dialogue to investigate what kinds of cognitive activities are activated during the role-play task. As a result, the learners collaboratively generated and monitored language and content for role play. Also, in order to accomplish tasks more successfully, learners shared the same understanding about the goal of the task, and tried to manage the task procedure. Through learner-learner dialogue, learners can participate in cognitive activities such as content, language construction, and task management voluntarily without the help from teachers. This means that learner-learner dialogue can be an activity to support language learning tasks. Also, it can make learners actively involved in learning and by sharing resources with each other. It is also important that learners can experience language use that participates in real-world communication activities, such as learning in the classroom and collaborating with peer learners. This study is an exploratory study for a basic understanding of learner's conversation as a cognitive activity, and the scope of the study is limited to clarifying contents of learner-learner dialogue as a cognitive activity in speaking tasks. Based on the findings of this study, future research should be conducted on the function of learner-learner dialogue as a cognitive activity in Korean language learning and its role in the classroom of Korean language education.