• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1211-1212
• /
• 2023

현재 대부분의 국내 학술 데이터 베이스는 개별 학술지 논문의 주제를 파악하는 표준화된 정보를 거의 제공하지 않고 있다. 본 연구에서는 논문의 제목만을 활용하여 학술 논문의 분야를 자동으로 분류하는 방법을 제안한다. 이를 위해 한국어로 사전 훈련된 KLUE-RoBERTa 모델을 사용하며, Back Translation 과 Chat-GPT 를 활용한 데이터 증강을 통해 모델의 성능을 향상한다. 연구 결과, Back Translation 과 Chat-GPT 를 사용하여 증강한 모델이 원본 데이터를 학습한 모델보다 약 11%의 성능 향상을 보였다.

• Journal of the Korean Chemical Society
• /
• v.46 no.1
• /
• pp.90-96
• /
• 2002

The STS contents, emphasized in the 6th curriculum, in the chemistry textbooks(II) were analyzed. The STS contents in textbooks showed average value of 2.7%. The chapter of ‘chemical bond and compound' were included 3.8% of STS contents. And the chapter of ‘atomic structure and periodic table', ‘state of material and solution', ‘science of material', and ‘chemical reaction' contained 3.2%, 2.2%, 1.9%, and 1.9% of STS contents, respectively. When the STS contents were analyzed by STS topics of Piel, the results are as follows; 33.7% on effect of technological developments, 27.5% on environmental quality and utilization of natural source, 19.6% on human engineering, 13.8% on energy, and 5.4% on sociality of science. However, there were no topics on population, space research and national defense. When the STS contents were analyzed by student activities of SATIS, most of the activities were research and case study. There were few field activities of practical investigation, problem solving and decision making, research design and stimulation. There were no activities of role play.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.38 no.3
• /
• pp.41-52
• /
• 2010

The purpose of this paper is to evaluate visitor satisfaction(VS) in waterfront parks in terms of resources, facilities, embodiment of theme(ET), site composition(SC), relaxation activity space(RAS), and dynamic activity space(DAS), which are supply-side components in the planning process of waterfront parks, in order to answer the research question; 'How is visitor satisfaction of waterfront parks determined?' After reviewing the literature on parks and the building process of waterfront parks in Korea, we constructed a conceptual framework and have ascertained a research hypothesis. We had obtained data through a questionnaire survey from 327 visitors at waterfront parks, based on the quota sampling method. We have analyzed the data using the path analysis method. We found that: 1) The direct effects of resources and facilities on VS turned out to be 0.273 and 0.306, respectively while the indirect effects are 0.114, 0.170, respectively. 2) The direct effects of SC, as a component of the planning process on VS, turned out to be 0.243 while that of ET had no affect on VS. The indirect effect of ET and SC on VS turned out to be 0.059 and 0.018, respectively. 3) The direct effects of RAS on VS turned out to be 0.129 while the indirect effects of RAS and DAS on VS turned out to be 0.002 and 0.017, respectively. 4) The size of causal effect, in order, were facilities, resources, SC, RAS, ET, and DAS. 5) Resources and facilities, as a park foundation, compose 64.84 percent of total causal effect while ET and SC have 24.04 percent and RAS and DAS have 11.12 percent, respectively. These research results imply that: 1) Existing waterfront parks should be regenerated with the embodiment of water related theme and with improved facilities for RAS and visitor programs and/or facilities for DAS. 2) The relationship among ET, SC, RAS and DAS should be increased for a significant improvement of VS, and 3) A process-oriented approach turned out to be highly useful for the development of substantive theory and methodology. It is recommended that a structural equation model on waterfront parks be developed using more empirical data and this approach be widely applied for testing its validity.

• Journal of Korean Home Economics Education Association
• /
• v.29 no.1
• /
• pp.71-92
• /
• 2017

The purpose of the present study is to investigate the experiences of original family and social relationships in single women. Semi-structured in-depth interviews were arranged with 18 single women who met the criteria for the present research purpose. We considered only those responses of the interviewees that are directly related to their experiences of original family and social relationships, where the collected data were analyzed in three steps by methods of thematic analysis. As a result of the analysis, the experiences of original family and social relationships in single women could be grouped into six main clusters of themes which may be labeled as (1) themes centered on 'Family attachment', (2) themes centered on 'Special People' and (3) themes centered on 'Estrange relative relationship'. (4) themes centered on 'Small social life', (5) themes centered on 'Absence of a social relationshipship', (6) themes centered on 'Drive for freedom'. Among the first category of themes of 'Family attachment', we have found three sub-themes such as 'Adult Baby under parents' care', 'Confirm of single life', 'Intricate family relations'. As for the second category of themes 'Special People', three sub-themes have been found including 'I Only Love Mom.', 'Sister friend', 'My lovely nephew or niece'. And for the third category of only one theme 'Estranged relative relationship'. And for the fourth category of themes 'Small social life', there were found four sub-themes which include 'Relationship at work', 'Chat with friends', 'Keep oneself aloof from the opposite sex', 'Singles gather together'. And for the fifth category of themes 'Absence of a social relationshipship', there were found three sub-themes which include 'Alone than meeting', 'Lazism', 'Tight daily lives'. And for the final category of themes 'Drive for freedom', there were found two sub-themes which include 'Caution for attachment', 'Caution for entanglement'. In all, there are sixteen sub-themes to consider. On the basis of these results, we presented some conclusions on the experiences of original family and social relationships in single women. We also presented some implications of these results on population education and future research.

• The KIPS Transactions:PartA
• /
• v.11A no.3
• /
• pp.217-226
• /
• 2004

This paper describes an experiment of concrete and specific group work learning system based on the traditional Jigsaw group work learning model. Jigsaw model has two groups of students such as random group and expert group so that a course can make progress on explaining and lecturing all members of class after each student can be a member of expert group of course topic. We design and implement Web-based training system to support collaboration and Interaction among students of a course based on Jigsaw model The Web- based learning system makes each group going up to the expert level of a course subject by supporting various study menu and provides equal opportunity of improving social abilities such as leadership, communication skill, trust, and trouble-settling by taking part in collaboration activities.

• Journal of Korean Library and Information Science Society
• /
• v.39 no.2
• /
• pp.493-514
• /
• 2008

The paper analyzes local children's libraries' status of application of classification scheme and problems and difficulties in its application in order to be used as basic material for development of classification scheme for exclusive use of children's libraries. Through survey, the paper researched 96 local children's libraries, and the results showed that KDC-applied libraries numbered 53 and were the highest, libraries which developed their own classification scheme numbered 22, and libraries which applied classification scheme for children's library numbered 13. Further, many asserted that there needs to be a development of standard classification scheme to be used in children's library. On problems and difficulties in applying classification scheme, notes such as difficulty of accommodating various forms of children's books(such as comic books and picture books), placing too much emphasis on collection of books on certain topic, and lack of topic categories appropriate for characteristics of collected books were voiced, and by holding additional discussion on these, the paper raised the usefulness of the dissertation to be used as basic materials for developing classification scheme for exclusive use of children's library in the future.

• The KIPS Transactions:PartD
• /
• v.19D no.2
• /
• pp.133-138
• /
• 2012

Queries entered in a search box are the results of users' activities to actively seek information. Therefore, search logs are important data which represent users' information needs. The purpose of this study is to examine if there is a relationship between the results of queries automatically classified and the categories of documents accessed. Search sessions were identified in 2009 NDSL(National Discovery for Science Leaders) log dataset of KISTI (Korea Institute of Science and Technology Information). Queries and items used were extracted by session. The queries were processed using an automatic classifier. The identified queries were then compared with the subject categories of items used. As a result, it was found that the average similarity was 58.8% for the automatic classification of the top 100 queries. Interestingly, this result is a numerical value lower than 76.8%, the result of search evaluated by experts. The reason for this difference explains that the terms used as queries are newly emerging as those of concern in other fields of research.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.4
• /
• pp.11-17
• /
• 2010

In this paper, we propose a recommendation system for supporting self-directed programming language education. The system is a recommendation system using collaborative filtering based on learners' level and stage. In this study, we design a recommendation system which uses collaborative filtering based on learners' profile of their level and correlation profile between learning topics in order to increase self-directed learning effects when students plan their learning process in e-learning environment. This system provides a way for solving a difficult problem, that is providing programming problems based on problem solving ability, in the programming language education system. As a result, it will contribute to improve the quality of education by providing appropriate programming problems in learner"s level and e-learning environment based on teaching and learning method to encourage self-directed learning.

• Journal of Korean Elementary Science Education
• /
• v.21 no.2
• /
• pp.289-296
• /
• 2002

The purpose of this study is to analyze STS contents in the elementary school science textbooks developed under the 7th national curriculum. Major findings from the analyses are as follows: (1) The Percentage of STS contents included is 26% by the number of pages, and 40% by the lesson hour, Higher proportions of the STS contents are found in the domains of ‘energy’ and ‘material’. By grade level, the coverage of 5th grade textbooks is relatively low. (2) Most STS topics are related to science rather than technology or society. (3) In many cases, STS contents are used in the whole processes of a lesson. (4) Most activities are 'group activity' or 'experiment.observation'. There are few activities of 'role playing', 'spot study' or 'interview'. These results indicate that recent STS education trends are reflected on the revised textbooks well. However, it is suggested that some improvement is needed, such as diversification of topic or activity.

• Journal of the Korean Society for information Management
• /
• v.25 no.4
• /
• pp.227-243
• /
• 2008

This study investigates an event detection method with the aim of generating an event-focused news summary from a set of news articles on a certain event using a multi-document summarization technique. The event detection method first classifies news articles into the event related topic categories by employing a SVM classifier and then creates event clusters containing news articles on an event by a modified single pass clustering algorithm. The clustering algorithm applies a time penalty function as well as cluster partitioning to enhance the clustering performance. It was found that the event detection method proposed in this study showed a satisfactory performance in terms of both the F-measure and the detection cost.