• Journal of the Korean Society for information Management
• /
• v.37 no.4
• /
• pp.1-26
• /
• 2020

In today's increasingly data-intensive academic environment, data is becoming the foundation of academic communication as a research outcome rather than a research by-product. However, there is a limit to guaranteeing actual data reuse only by expanding the data supply or securing accessibility. In order to overcome this, it is necessary to understand the data reuse behavior and data needs in-depth. Therefore, this study attempted to identify the major data reuse behavior and data needs among researchers. To this end, the authors of KCI papers among the data reuse documents of the Korea Social Science Data Archive (KOSSDA) for the past 3 years were targeted. An in-depth interview was conducted with 12 researchers who accepted the interview. As a result, factors considered when reusing data were personal, economic, technical, and social aspects, and it was found that the data itself was used or contextual information of the data was used depending on the purpose of data reuse. The path to acquiring data is a web-based source of information, and a path through informal communication can also be found. In terms of the data needs, it was found that they prefer English, the United States, and institutional producers. Also they have a clear preference for quantitative data from an interviewer-filled interpersonal interview survey method, rich metadata along with raw data, and data that contains identification information. However, due to the lack of confidence in the value, it is negative for the use of data with controlled access and use, and it is difficult to confirm a clear preference because there is no similar data available for selection in terms of size and freshness.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.378-390
• /
• 2022

With the recent development of deep composite multiplication neural network learning, deep learning techniques applied to single-image super-resolution have shown good results, and the strong expression ability of deep networks has enabled complex nonlinear mapping between low-resolution and high-resolution images. However, there are limitations in applying it to real-time or low-power devices with increasing parameters and computational amounts due to excessive use of composite multiplication neural networks. This paper uses blocks that extract hierarchical characteristics little by little using information distillation and suggests the Recursive Distillation Super Resolution Network (RDSRN), a lightweight network that improves performance by making more accurate high frequency components through high frequency residual purification blocks. It was confirmed that the proposed network restores images of similar quality compared to RDN, restores images 3.5 times faster with about 32 times fewer parameters and about 10 times less computation, and produces 0.16 dB better performance with about 2.2 times less parameters and 1.8 times faster processing time than the existing lightweight network CARN.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.8
• /
• pp.339-346
• /
• 2022

Recently, the resource depletion and climate change problem caused by the massive usage of fossil fuels for electric power generation has become a critical issue worldwide. According to this issue, interest in renewable energy resources that can replace fossil fuels is increasing. Especially, photovoltaic power has gaining much attention because there is no risk of resource exhaustion compared to other energy resources and there are low restrictions on installation of photovoltaic system. In order to use the power generated by the photovoltaic system efficiently, a more accurate photovoltaic power forecasting model is required. So far, even though many machine learning and deep learning-based photovoltaic power forecasting models have been proposed, they showed limited success in terms of interpretability. Deep learning-based forecasting models have the disadvantage of being difficult to explain how the forecasting results are derived. To solve this problem, many studies are being conducted on explainable artificial intelligence technique. The reliability of the model can be secured if it is possible to interpret how the model derives the results. Also, the model can be improved to increase the forecasting accuracy based on the analysis results. Therefore, in this paper, we propose an explainable photovoltaic power forecasting scheme based on BiLSTM (Bidirectional Long Short-Term Memory) and SHAP (SHapley Additive exPlanations).

• Journal of Korean Society of Forest Science
• /
• v.111 no.1
• /
• pp.165-176
• /
• 2022

To raise awareness of carbon reduction in climate change, the Korea Forest Service has developed and adopted a forest carbon offset program, which aims to reduce carbon levels based on forest management. However, to maintain the forest carbon offset program, challenges such as the lack of a forest monitoring system to manage and maintain the program, must be faced. In this context, we investigated the limitations of conducting forest carbon offset programs using a number of interview techniques, including in-depth interview and questionnaire survey methods. The questionnaire surveys were developed based on the results of a literature review along with a preinterview and in-depth survey of the people in charge of the forest carbon offset program. The Irving Seidman technique was adopted for the in-depth interviews. Additionally, descriptive and frequency analyses were conducted to identify the characteristics of perception. Lastly, logistic regression was used to identify the limiting factors that affect the willingness to perform forest carbon offset monitoring activity. Results showed that the project managers or people in charge of the forest carbon offset program lacked expertise in forest carbon offset programs, which negatively affected their willingness to perform monitoring activity. Additionally, the study revealed a number of limiting factors that hindered the monitoring of forest carbon offset projects. Improving understanding using the approaches presented in this study may contribute to increasing the benefits associated with the forest carbon offset program in South Korea.

• Journal of Korean Home Economics Education Association
• /
• v.36 no.2
• /
• pp.101-117
• /
• 2024

The aim of this study is to identify the competencies required for fashion-related majors that meet contemporary demands, align with the objectives of university education, and reflect the qualities desired in graduates. To achieve this goal, we conducted content analysis of relevant data and in-depth interviews with experts. First, the content analysis involved coding key information from the introductions, educational goals, desired qualities of graduates, and curricula published on the websites of both South Korea and international fashion-related universities. Additionally, we analyzed the National Competency Standards (NCS) and the Meta-goals of higher education programs set by the International Textile Apparel Association (ITAA), extracting six core competencies. Second, in-depth interviews were conducted with six experts, each with 23 to 31 years of experience in Korean and international apparel industry and academia. The interviews were recorded, transcribed, and keywords were extracted. To ensure the validity of the coding results, cross-checks were performed among the researchers. The analysis identified the following competencies: empathic communication, social responsibility, professional thinking, creative and integrative thinking, global perspective, and challenging leadership. Based on these findings, establishing competencies that meet contemporary demands and developing corresponding curricula are essential steps towards creating a feedback system. Future research should focus on developing and implementing curricula that foster a virtuous cycle, ultimately enhancing students' competency levels.

• The Journal of Korean Association of Computer Education
• /
• v.17 no.2
• /
• pp.21-29
• /
• 2014

In spite of the R&D level of Korea, the efforts to protect the R&D results from outflowing has not been raised up. We investigated the current status of security education and the level of researcher's awareness for research security in the government-financed institutes. Also, we attempted to find out the needs for institutionalization of the security education. We conducted a survey and in-depth interviews of all the security officers in the thirty-seven government-financed institutes. The results show that the awareness level of the researchers for R&D security is below adequate level, and that security education is necessary in order to increase the security awareness. Also, it is necessary to institutionalize the security education.

• 한국신재생에너지학회:학술대회논문집
• /
• 2010.06a
• /
• pp.192.2-192.2
• /
• 2010

일반적 의미의 소수력발전은 계곡이나 저낙차의 하천에서 시도되었으나, 한국의 지형과 강수패턴등은 소수력발전을 활성화하기에 어려운 점들이 있었다. 이에 최근에는 정수장, 하수처리장등과 같은 인공구조물에 소수력발전을 설치 운영하는 방향으로 가고 있으며, 특히 화력발전소 냉각공정에 사용되는 해수를 이용한 소수력발전이 크게 성공하였고 확대설치 되어가고 있다. 해안에 위치하는 LNG인수기지에서는 LNG의 기화에 해수를 열원으로 사용하며, 기화공정에서 열교환 후 바다에 배출된다. 이 때 기화해수와 공기와의 접촉으로 생성된 거품은 해양미생물과의 복합작용으로 쉽게 깨어지지 않고 바다로 떠내려가게 된다. 이러한 거품은 시각적 거부감으로 인하여 인근어민들의 불편함을 야기하고 있으며, 또한 배출해수와 일반해수와의 온도차로 인한 인근 어장이나 양식장의 어획고에 미칠 수 있는 부작용의 가능성에 대한 우려는 더욱 방류해수의 적절한 처리를 필요로 하고 있다. 이러한 방류해수의 거품생성을 해결하는 데 있어 근본적인 해결방법은 심층배수법인데, 심층배수 구조물에 발전수차를 추가 설치만 하면 수력발전이 가능하다. 방류해수의 거품관련 환경문제를 해결하면서 동시에 청정전력을 생산할 수 있는 해양소수력발전에 대하여 KOGAS에서는 LNG 인수기지에의 적용가능성을 분석하고 있으며, 방류해수의 낙차와 조수간만의 차를 이용하는 해양소수력발전을 LNG 인수기지에의 적용하는 것으로는 세계최초의 시도이다. 주변지형에 따른 입지여건을 분석하고, 해수계통분석, 소수력발전방법, 수차종류, 수차용량, 수차개수, pond의 크기등을 결정하고, 수리해석 및 경제성분석을 수행할 것이며 소수력발전의 타당성여부에 대한 가늠을 잡고자 한다.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.45 no.6
• /
• pp.77-81
• /
• 2008

In the deeply coupled GPS/INS integrated systems, if the integration filter update period is longer than the period of GPS navigation data, the loss of correlation values occurs due to the bit transition. This problem can be resolved when data wipe off(DWO) is used. However, general DWO methods requires heavy computation or cannot be applied continuously. This paper proposes an effective DWO method using carrier phase discriminator In order to show validity of the proposed method, simulations were carried out. The simulation results show that the data bit is accurately estimated and conform that the loss of correlation values and the error of code phase is small.

• Journal of Korean Library and Information Science Society
• /
• v.48 no.3
• /
• pp.153-176
• /
• 2017

This article or study subjectively explores actual experiences of a human book from a phenomenal viewpoint and is based on considerable amounts of reference literature and the collecting/analyzing of precedent theses data. Selecting specific research samples and conducting in-depth interviews were made to elicit the phenomenal marks or responses in human book experiences. Samples included a total of three human book practicing teachers belonging to Gyonggi Provincial Office of Education, who were picked out or chosen according to their rich performance experiences/activities over a two year period. They were interviewed in four different basic topic areas or parts sectioned out to the following: 1) actual needs of a human book; 2) experiences in human book; 3) changes of cognition to the human book; and, lastly 4) improvement in the application of a human book. As a result, the number of schools, the number of readers, and the number of people who participated in human book activities continued to expand. In addition, human book activities have become an opportunity for growth not only for participating readers, but also for the human book itself.

• Journal of Broadcast Engineering
• /
• v.21 no.6
• /
• pp.977-985
• /
• 2016

In this paper, we propose a new method for on-line genre classification using spectrogram and deep neural network. For on-line processing, the proposed method inputs an audio signal for a time period of 1sec and classifies its genre among 3 genres of speech, music, and effect. In order to provide the generality of processing, it uses the spectrogram as a feature vector, instead of MFCC which has been widely used for audio analysis. We measure the performance of genre classification using real TV audio signals, and confirm that the proposed method has better performance than the conventional method for all genres. In particular, it decreases the rate of classification error between music and effect, which often occurs in the conventional method.