• Journal of the Korean Society for Library and Information Science
• /
• v.43 no.1
• /
• pp.313-332
• /
• 2009

Cross-language text categorization(CLTC) can classify documents automatically using training set from other language. In this study, collections appropriated for CLTC were extracted from KTSET. Classification performance of various CLTC methods were compared by SVM classifier using machine translation. Results showed that the classification performance in the order of poly-lingual training method, training-set translation and test-set translation. However, training-set translation could be regarded as the most useful method among CLTC, because it was efficient for machine translation and easily adapted to general environment. On the other hand, low performance was shown to be due to the feature reduction or features with no subject characteristics, which occurred in the process of machine translation of CLTC.

• The Journal of the Korea Contents Association
• /
• v.18 no.6
• /
• pp.50-62
• /
• 2018

This study is to ascertain the meaning of personal media as cultural intermediaries in the cultural diffusion, acceptance, and intermediation for the production and consumption of cultural contents. We examined the case of personal media, , which is one of popular YouTube channel in Korea and analyzed fifty most viewed videos in last two years based on four perspectives of digital cultural intermediaries: forms, expression methods, linguistic features, and evaluations. As expected, results demonstrate that shows characteristics of digital intermediary: diverse subjects without fixed rules, a variety of multimedia effects, colloquial style including neologism and multilingual, and users' high preferences caused by creator's sincerity. Our findings and limitations can serve directions of further empirical researches.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.22 no.4
• /
• pp.45-70
• /
• 2011

Today's S. Korean society has become more multicultural, so many libraries in S. Korea have been working hard to serve their "new customers". In this regard, this study intends to investigate library cooperative systems for acquisition of multicutural materials in Western countries to relieve difficulties of libraries in S. Korea in acquiring multicutural materials. the study analyzed 23 national libraries, state libraries, public libraries and library systems in North America, Australia and Northern European Countries. The research found categorization schemes. First, library cooperative systems are divided into two groups according to cooperative methods: cooperative acquisition and bulk loan method. Second, the systems are categorized according to the territorial scope of the service: country, state or province, borough or city and inter-library. The study also analyzed and synthesized the characteristics of these systems. It finally provides recommendations to apply the methods used by the library cooperative systems in Western countries to Korean's situation.

• Korean Journal of Health Education and Promotion
• /
• v.31 no.2
• /
• pp.65-78
• /
• 2014

Objectives: The aim of this study is to evaluate promotional Internet health information on health functional foods with regard to labeling information, website structure, and overall quality. Methods : The websites of 15 producers of three health functional foods (Red ginseng, vitamin, aloe) were selected. Ministry of Food and Drug Safety (MFDS) required labeling information was used to evaluate the integrity of basic information, and the structural properties of the websites were investigated. Moreover, DISCERN instrument was used to evaluate reliability and quality of information. Results : 1) Among MFDS required labeling information, seven items, including 'sell by date', and 'possible side effects' were not fully specified, and only 6.7% of Internet advertisements provided 'possible side effects'. 2) Each of 92.9% of these websites offered 'Introducation to websites and producers', and 'customer service', whereas only 64.3% and 42.9% of these websites offered 'FAQ about website utilization' and 'multilingual support', respectively. 3) The evaluation using DISCERN instrument showed the scores of $2.03{\pm}0.24$ in reliability, $1.87{\pm}0.28$ in quality, and $2.10{\pm}0.55$ in overall quality. Conclusions : Current promotional Internet health information on health functional foods has various weaknesses in terms of reliability and quality that need to be improved.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.176-179
• /
• 2019

최근 세계적으로 한국에 대한 인식이 좋아짐으로써 한국으로 유학이나 여행을 오는 외국인이 많아지고 있다. 외국은 다양한 문화와 언어를 가지고 있고 그 외국의 문화와 언어에 익숙하지 않은 많은 식당에서 그들은 식사함에 어려움을 겪고 있다. 우리 학교의 경우에서도 마찬가지로 유학 와 있는 한국어가 서툰 학생들이 한국에서 가장 불편한 일 중 하나가 식당에서 음식을 시키는 것이라고 할 정도로 식사가 제한되어 있다. 이를 해결하기 위해 본 논문에서는 QR 코드를 메뉴판에 부착하여 그 QR 코드를 찍기만 하면 바로 원하는 언어를 선택할 수 있고, 그들의 언어로 음식에 대한 설명이 나타나는 시스템을 제안하고 개발하였다. 제안한 시스템은 종교에 따라 먹지 못하거나, 식습관에 따라 먹지 않는 음식이 있는 외국인들 역시 전자 메뉴판을 이용함으로 어려움을 해결 할 수 있다. 제안한 시스템은 그들 주변에 어떤 식당이 있고 어떤 음식이 있는지 쉽게 알 수 있고 그 어플을 사용하는 사람들과 소통할 수 있는 어플리케이션을 제작함으로써 유학생뿐만 아닌 일반 관광객들을 대상으로도 적극적으로 활용 할 수 있을 것으로 기대가 된다.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.179-186
• /
• 2021

We have been developed the smart tourism app and the Instagram and YouTube contents to provide personalized tourism information and travel product information to individual tourists. In this paper, we develop a rule-based smart tourism chatbot with the khaiii (Kakao Hangul Analyzer III) morphological analyzer and Neo4J graph database. In the proposed chatbot system, we use a morpheme analyzer, a proper noun dictionary including tourist destination names, and a general noun dictionary including containing frequently used words in tourist information search to understand the intention of the user's question. The tourism knowledge base built using the Neo4J graph database provides adequate answers to tourists' questions. In this paper, the nodes of Neo4J are Area based on tourist destination address, Contents with property of tourist information, and Service including service attribute data frequently used for search. A Neo4J query is created based on the result of analyzing the intention of a tourist's question with the property of nodes and relationships in Neo4J database. An answer to the question is made by searching in the tourism knowledge base. In this paper, we create the tourism knowledge base using more than 1300 Jeju tourism information used in the smart tourism app. We plan to develop a multilingual smart tour chatbot using the named entity recognition (NER), intention classification using conditional random field(CRF), and transfer learning using the pretrained language models.

• The Bulletin of The Korean Astronomical Society
• /
• v.43 no.2
• /
• pp.49.3-50
• /
• 2018

Korea has kept its records of astronomical phenomena since around 2,000 years ago. However, the contents and scope of relevant service have been limited for researchers who need those records due to lack of complete data collection. In this regard, it is necessary to establish efficient collection and management systems of Korean astronomical records by utilizing an environment that is easily accessible. This study is intended to complete the development of a testbed system that allows researchers to systematically input and validate, in a Web environment, multiple astronomical records among the historical documents until Modern Joseon after the Three Kingdoms Period. Recognition of the pre-translated data and tables in advance is followed by its storage in the database built on the Web. Then, data validation is implemented by providing a retrieval service according to a specific form to only a finite number of researchers who have access authority. This study is targeted at a testbed system that takes around three months to be completely developed. The completed testbed system is expected to allow internal and external researchers of an organization to easily access the service on the Web. This will ensure that the accuracy of the data can be verified mutually and help identify areas of service improvement. The opinions collected regarding service improvement will be reflected in the future system. Eventually, domestic astronomical records will subsequently be able to be utilized internationally through the multilingual service.

• International Journal of Computer Science & Network Security
• /
• v.21 no.7
• /
• pp.63-69
• /
• 2021

Flipped learning is necessary for modern education but quite difficult to implement. In pedagogical science, the question remains to what extent the practical work of the teacher in combination with the technologies of flipped learning will improve the quality of higher education. The aim of this article is to study the effectiveness and feasibility of using flipped learning technologies, assessing their perception by students (advantages and problems), identified an algorithm for introducing flipped learning technology in higher education institutions. Research methods. The main method is an experiment. An evaluation of the effectiveness of the study was conducted using a questionnaire and observation method. Statistical methods were used to evaluate the results of the experiment. The research hypothesis is that flipped learning allows the teacher to spend more time on an individual approach, to understand the real needs of students, and provide effective feedback, thereby improving the quality of learning and motivation of students, especially while studying complex material. The results of the study are to prove the effectiveness of the technology of flipped education in the study of complex disciplines, courses, topics. The use of flipped learning strategies improves the self-regulation of the educational process, group work skills, improves students' ability to learn, overcome difficulties. The technology of flipped learning in the presence of modern technical means and constant work on improving the level of digital literacy is an effective means for students to master complex topics and problematic issues that require additional consideration and discussion. The perspective of further research is the consideration of integrated approaches to the application of flipped learning technologies to the principles of STEAM-education, multilingual and multicultural programs, etc. It is also worth continuing to develop a set of methods aimed at enhancing the student's learning activities, the formation of group work skills, direct participation in creating the foundations of higher education.

• Journal of the Korea Convergence Society
• /
• v.13 no.2
• /
• pp.21-29
• /
• 2022

Based on the importance of dialect research, preservation, and communication, this paper conducted a study on machine translation of Korean dialects for dialect users who may be marginalized. For the dialect data used, AIHUB dialect data distributed based on the highest administrative district was used. We propose a many-to-one dialect machine translation that promotes the efficiency of model distribution and modeling research to improve the performance of the dialect machine translation by applying Copy mechanism. This paper evaluates the performance of the one-to-one model and the many-to-one model as a BLEU score, and analyzes the performance of the many-to-one model in the Korean dialect from a linguistic perspective. The performance improvement of the one-to-one machine translation by applying the methodology proposed in this paper and the significant high performance of the many-to-one machine translation were derived.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.13 no.2
• /
• pp.102-110
• /
• 2017

Current word sense disambiguation techniques employ various machine learning-based methods. Various approaches have been proposed to address this problem, including the knowledge base approach. This approach defines the sense of an ambiguous word in accordance with knowledge base information with no training corpus. In unsupervised learning techniques that use a knowledge base approach, graph-based and similarity-based methods have been the main research areas. The graph-based method has the advantage of constructing a semantic graph that delineates all paths between different senses that an ambiguous word may have. However, unnecessary semantic paths may be introduced, thereby increasing the risk of errors. To solve this problem and construct a fine-grained graph, in this paper, we propose a model that iteratively constructs the graph while eliminating unnecessary nodes and edges, i.e., senses and semantic paths. The hybrid similarity estimation model was applied to estimate a more accurate sense in the constructed semantic graph. Because the proposed model uses BabelNet, a multilingual lexical knowledge base, the model is not limited to a specific language.