• ETRI Journal
• /
• 제46권1호
• /
• pp.59-70
• /
• 2024

We introduce a high-performance named entity recognition (NER) model for written and spoken language. To overcome challenges related to labeled data scarcity and domain shifts, we use transfer learning to leverage our previously developed KorBERT as the base model. We also adopt a meta-pseudo-label method using a teacher/student framework with labeled and unlabeled data. Our model presents two modifications. First, the student model is updated with an average loss from both human- and pseudo-labeled data. Second, the influence of noisy pseudo-labeled data is mitigated by considering feedback scores and updating the teacher model only when below a threshold (0.0005). We achieve the target NER performance in the spoken language domain and improve that in the written language domain by proposing a straightforward rollback method that reverts to the best model based on scarce human-labeled data. Further improvement is achieved by adjusting the label vector weights in the named entity dictionary.

• 영어어문교육
• /
• 제16권1호
• /
• pp.19-40
• /
• 2009

Teacher written feedback on student compositions has received tremendous attention in second language (L2) writing research. Notwithstanding the importance of understanding both teachers' and students' perspectives on the feedback process, much of the feedback research has only looked into one-side of the story - adopting either the teacher's or the student's perspective. The current study is an attempt to look into both sides of the story by examining the types of written feedback that students prefer, the extent to which students' preferences and teachers' actual feedback practice overlap, and the extent to which student perceptions of teacher feedback coordinate teacher self-reflections on their feedback practice. Three English composition classes (3 teachers and 46 students) at a university participated in this study. It analyzed student and teacher data from questionnaires and teacher written feedback on student compositions. The results showed that students' preference for feedback on global and local issues varied across the three composition classes. This is partly a consequence of how students perceived the type of feedback that their teachers practiced. Teacher self-reflection on and student perception of teacher written-feedback generally coordinated. These findings are discussed in light of how contextual factors affect learner perception of teacher written feedback and underscore the need for examining students' reactions to feedback and teacher self-reflection.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2017년도 제29회 한글 및 한국어 정보처리 학술대회
• /
• pp.260-263
• /
• 2017

This paper is to broaden the possible spectrums of analyzing the Korean-written novel "The Vegetarian" by using the computational linguistics program. Through the use of language model, which was usually used in bi-gram analysis in corpus linguistics, to the International Man Booker award winning novel, the characteristics of "The Vegetarian" is investigated by comparing it to the English-written novel "A Little Life".

• International Journal of CAD/CAM
• /
• 제5권1호
• /
• pp.83-90
• /
• 2005

This is a report on a new methodology to manage design knowledge by utilizing a knowledge-based CAD and a prototype system named $C^3$ (Cubic; CAD knowledge Code Capacitor), which is being developed using our methodology. $C^3$ facilitates (i) the automatic generation of a knowledge code for a knowledge-based CAD by processing design documents written in the format near the natural language, such as English or Japanese, and (ii) automatically generation of a design document written in the format near the natural language from the knowledge code. The features of the system facilitate document-based design knowledge management which reduces the designer's load to encode and maintain design knowledge, because it is easier for a designer to treat a natural language description than a coded description.

• 인문언어
• /
• 제2권2호
• /
• pp.143-161
• /
• 2002

This essay explores the theories of Italian poet and filmmaker Pier Paolo Pasolini on the language of cinema. In essays such as "The Cinema of Poetry" and "The Written Language of Reality" composed during the 1960s, Pasolini argues for the special status of film language as "pre-grammatical" and links it to visual signifying processes such as dreams and memories. He also views cinema as the inroads towards a general semiotics of reality since, for him, the basic unit of film language is not the shot but those objects of reality that constitute the mise-en-scene of the shot, hence cinema is posited as the written language of reality whose minimal units of articulation are the very objects of reality itself. Accused by semioticians such as Umberto Eco of semiotic ingenuousness in trying to reduce the facts of culture to nature, Pasolini responded by arguing that he was trying to do the opposite, that is to say, to culturalize nature by examining it as a language. Against the constructed naturalism of both commercial and neorealist films, Pasolini argued for the creation of a poetic cinema able to exploit its constitutional pre grammatical, oneiric and sacred relationship with the world. The essay concludes with an analysis of the film Medea in which Pasolini′s attempt to restore a sacred vision of reality merges with his concerns over the cultural genocide of traditional and emarginated peoples at the hands of neocapitalist homologation.

• 한국멀티미디어학회논문지
• /
• 제7권6호
• /
• pp.824-831
• /
• 2004

This paper presents the java bytecode-to-.NET MSIL intermediate language translator which enables the execution of the java program in .NET environments without JVM(java Virtual Machine), translating bytecodes produced by compiling java programs into MSIL codes. Java, one of the most widely used programming languages recently, is the language invented by James Gosling at Sun Microsystems, which is the next generation language independent of operating systems and hardware platforms. Java source code is compiled into bytecode as intermediate code independent of each platform by compiler, and also executed by JVM. .NET language such as C# and .NET platform in Microsoft Corp. has been developed to meet the needs of programmers, and cope with Java and JVM platform of Sun Microsystems. After compiling, a program written in .NET language is converted to MSIL code, and also executed by .NET platform but not in JVM platform. For this reason, we designed and implemented the java bytecode-to-.NET MSIL translator system for programs written in java language to be executed in the. NET platform without JVM. This work improves the execution speed of programs, enhances the productivity, and provides a environment for programmers to develop application programs without limitations of programming languages.

• 한국어정보학회:학술대회논문집
• /
• 한국어정보학회 2017년도 제29회 한글및한국어정보처리학술대회
• /
• pp.260-263
• /
• 2017

This paper is to broaden the possible spectrums of analyzing the Korean-written novel "The Vegetarian" by using the computational linguistics program. Through the use of language model, which was usually used in bi-gram analysis in corpus linguistics, to the International Man Booker award winning novel, the characteristics of "The Vegetarian" is investigated by comparing it to the English-written novel "A Little Life".

• 비교문화연구
• /
• 제24권
• /
• pp.170-183
• /
• 2011

This study explored how a Chinese college student who previously had not reached a threshold level of Korean proficiency used L1 (Chinese) and L2 (English) as a tool to socialize into Korean (L3) culture of learning over the course of study. From a perspective of language socialization, this study examined the cross-linguistic influence of L1 and L2 on the L3 acquisition process by tracing an approach to language learning and practices taken by the Chinese student as a case study. Data were collected through three methods; interview protocols, various types of written texts, and observations. The results showed that the student used English as a means to negotiate difficulties and expertise by empowering her L2 exposure during the classroom practices. Her ways of using L2 in oral practices could be characterized as the 'Inverse U-shape' pattern, under which she increased L2 exposure at the early stage of the study and shifted the intermediate language to L3 at the later stage of the study. When it comes to the language use in written practices, the sequence of "L2-L1-L3" use gradually changed to the "L2-L3" sequence over time, signifying the importance of interaction between L2 and L3. However, the use of her native language (L1) in a Korean-speaking classroom was limited to a certain aspect of literacy practices (i.e., vocabulary learning or translation). This study argues for L2 communication channel in cross-cultural classrooms as a key factor to determine sustainable learning growth.

• 디자인학연구
• /
• 제19권2호
• /
• pp.91-98
• /
• 2006

제품의 융합과 그 사용 환경의 복잡화로 인하여 다양한 감각을 활용해서 제품과 교감하는 다중 인터페이스의 필요성이 제기되고 있다. 전통적이고 일반적인 정보의 습득에는 시각이 압도적으로 많이 활용되었으나 디지털 네트워크 기술을 기반으로 발달할 미래 정보 사회에서는 보다 편리하고 합리적인 제품의 사용을 위하여 다양한 감각의 활용이 요구될 것이다. 정보 전달에 있어서 시각과 더불어 점차적으로 활용 가능성이 높아지고 있는 청각은 그 쓰임새가 더욱 넓어지고 또한 다양한 방식으로 발전될 것이다. 이러한 배경을 바탕으로 본 연구에서는 정보의 전달에서 사용되는 문자언어와 음성언어의 특성과 그에 대한 반응에 있어서 남성과 여성의 차이점의 비교분석에 관한 연구가 수행되었다. 이를 위하여 일단 언어를 구성하는 여러 가지 요소에 대한 문헌자료의 연구를 수행하였다. 이어서 시각과 청각의 특성을 검토한 후 선행 연구를 기반으로 하여 적절한 실험이 설계 및 수행되었다. 수행된 실험은 객관적인 분석방법을 통하여 실험 결과가 검토되었다. 금번 연구의 결과로는 첫째, 문자언어에 대한 반응 시간이 음성언어에 대한 그것보다 짧다는 것, 둘째 각 자극에 대한 반응에 있어서 남녀간의 차이가 존재한다는 것, 셋째, 시각이 청각에 비하여 선택적인 측면에서 절대적 우위를 점하지 못한다는 점등을 들 수 있다. 이번 연구를 기반으로 하여 보다 다양한 감각에 대한 폭 넓은 연구 방법의 개발이 필요할 것으로 사료된다.

• 한국어교육
• /
• 제28권4호
• /
• pp.1-38
• /
• 2017

This longitudinal study analyzed syntactic development (Complexity, Accuracy, and Fluency; CAF) of six L2 Korean learners in a natural learning context. The learners recalled the stories of a short animated video through speaking and writing every 3 months, from month 0 to 15. The learners' responses were analyzed for a series of CAF measures and their cognitive, psychological, and social variables were investigated. The results showed that (i) L2 Korean learners' speaking and writing in various time periods showed significant differences in spoken and written accuracy, and complexity; (ii) the correlation between spoken and written complexity, spoken and written accuracy, as well as spoken and written fluency were significant, and (iii) the regression analysis showed that learners' cognitive, social, and psychological variables have significant effect on the L2 Korean syntactic development. The current study reports that L2 Korean learners engaged in self-learning in a natural learning environment without formal instruction made significant syntactic development.