• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.39-48
• /
• 2023

Because a railway bridge is designed over a long period of time and covers a large site, it involves various environmental factors and uncertainties. For this reason, design changes often occur, even if the design was thoroughly reviewed in the initial design stage. In particular, design changes of large-scale facilities, such as railway bridges, consume significant time and cost, and it is extremely inefficient to repeat all the procedures each time. In this study, a technique that can improve the efficiency of learning after design change was developed by utilizing the learning result before design change through transfer learning among deep-learning algorithms. For analysis, scenarios were created, and a database was built using a previously developed railway bridge deep-learning-based prediction system. The proposed method results in similar accuracy when learning only 1000 data points in the new domain compared with the 8000 data points used for learning in the old domain before the design change. Moreover, it was confirmed that it has a faster convergence speed.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.6
• /
• pp.1467-1482
• /
• 2017

Recently, Interests on The Fourth Industrial Revolution has been increased. In the manufacturing sector, the introduction of Smart Factory, which automates and intelligent all stages of manufacturing based on Cyber Physical System (CPS) technology, is spreading. The complexity and uncertainty of smart factories are likely to cause unexpected problems, which can lead to manufacturing process interruptions, malfunctions, and leakage of important information to the enterprise. It is emphasized that there is a need to perform systematic management by analyzing the threats to the Smart Factory. Therefore, this paper systematically identifies the threats using the STRIDE threat modeling technique using the data flow diagram of the overall production process procedure of Smart Factory. Then, using the Attack Tree, we analyze the risks and ultimately derive a checklist. The checklist provides quantitative data that can be used for future safety verification and security guideline production of Smart Factory.

• The Journal of Engineering Geology
• /
• v.7 no.1
• /
• pp.11-26
• /
• 1997

Through the Korea-Germany joint project, a novel method, that is, an optical Corescanner (Stereophotogramrnetry) of acquisition, evaluation and display of struct-ural elements of drill cores has heen realized. AH scannable cores can he digitally stored on a storage device(dat tape, optical disc etc.) and available for further an-alysis using diverse software facilities. The use of this scanning technique was d-emonstrated on the cores derived from the metamorphosed sedimentary rocks in our country. Recent studies showed a great potential of using the Corescanner with high resolution not only for avoiding ambiguities of drill log interpretation due to the capability of accurate quantative analysis of structural elements, but also for replacing the cores themselves as a data-base one via completely copying of the core outlook.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.2
• /
• pp.11-25
• /
• 2001

There are several algorithms for classification in modeling relations, patterns, and rules which exist in data. We learn to classify objects on the basis of instances presented to us, not by being given a set of classification rules. The Bayesian learning uses the probability distribution to express our knowledge about unknown parameters and update our knowledge by the law of probability as the evidence gathered from data. Also, the neural network models are designed for predicting an unknown category or quantity on the basis of known attributes by training. In this paper, we compare the misclassification error rates of Bayesian Neural Network method with those of other classification algorithms, CHAID, CART, and QUBST using several data sets.

• Journal of Energy Engineering
• /
• v.23 no.3
• /
• pp.27-35
• /
• 2014

The existing flow regime map on mixed convection in vertical cylinders was investigated through an analysis of original literatures and its re-formation. The original literatures related to the existing map were reviewed. Using the investigated data and heat transfer correlations, the map was redrawn independently, and compared with the existing map. The redrawn map showed that mixed convection regime was not curved lines but straight lines and the transition regime was unable to be reproduced. Unlike the existing map with a little data, there are lots of data in the redrawn map. The reviews revealed that the existing map used the data selectively among the experimental and theoretical results, and a detailed description for lines forming mixed convection and transition regime was not provided. While considerable studies on mixed convection have been performed since that of Metais and Eckert, the existing map has still been used as the best method to distinguish natural, forced and mixed convection regime.

• Korean Journal of Microbiology
• /
• v.54 no.3
• /
• pp.254-265
• /
• 2018

Avian influenza recently damaged the poultry industry, which suffered a huge economic loss reaching billions of U.S. dollars in South Korea. Transmission routes of the pathogens would help plan to control and limit the spread of the devastating biological tragedy. Phylogenetic analyses of pathogen's DNA sequences could sketch transmission trees relating hosts with directed edges. The last decade has seen the methodological development of inferring transmission trees using epidemiological as well as genetic data. Here, I reanalyzed the DNA sequence data that had originated in the highly pathogenic avian influenza H5N8 outbreak of South Korea in 2014. The H5N8 viruses spread geographically contiguously from the origin of the outbreak, Jeonbuk. The Jeonbuk origin viruses were known to spread to four provinces neighboring Jeonbuk. I estimated the transmission tree of the host domestic and migratory wild birds after combining multiple runs of Markov chain Monte Carlo using a Bayesian method for inferring transmission trees. The estimated transmission tree, albeit with a rather large uncertainty in the directed edges, showed that the viruses spread from Jeonbuk through Chungnam to Gyeonggi. Domestic birds of breeder or broiler ducks were estimated to appear to be at the terminal nodes of the transmission tree. This observation confirmed that migratory wild birds played an important role as one of the main infection mediators in the avian influenza H5N8 outbreak of South Korea in 2014.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.634-636
• /
• 2005

지능형 로봇 기술이 발전하면서 영상에서 장면을 이해하는 연구가 많은 관심을 받고 있으며, 최근에는 불확실한 환경에서도 좋은 성능을 발휘할 수 있는 확률적 접근 방법이 많이 연구되고 있다. 본 논문에서는 확률적 모델링이 가능한 베이지안 네트워크(BN)를 이용해서 장면 인식 추론 모듈을 설계하고, 실제 환경에서 얻어진 증거 및 베이지안 추론 결과를 결합하여 분류 성능을 향상시키기 위한 방법을 제안한다. 영상 정보는 시간에 대해 연속성을 가지고 있기 때문에, 증거 정보와 베이지안 추론 결과들을 적절히 결합하면 더 좋은 결과를 예상할 수 있으며, 본 논문에서는 확신 요소(Certainty Factor: CF) 분석에 의한 결합 방법을 사용하였다. 성능 평가 실험을 위해서 SET (Scale Invariant Feature Transform) 기법을 이용하여 물체 인식 처리를 수행하고, 여기서 얻어진 데이터를 베이지안 추론의 증거로 사용하였으며, 전문가의 CF 값 정의에 의한 베이지안 네트워크 설계 방법을 이용하였다.

• Journal of Internet Computing and Services
• /
• v.13 no.5
• /
• pp.21-31
• /
• 2012

Threat Evaluation(TE) which has air intelligence attained by identifying friend or foe evaluates the target's threat degree, so it provides information to Weapon Assignment(WA) step. Most of TE data are passed by sensor measured values, but existing techniques(fuzzy, bayesian network, and so on) have many weaknesses that erroneous linkages and missing data may fall into confusion in decision making. Therefore we need to efficient Threat Evaluation system that can refine various sensor data's linkages and calculate reliable threat values under unpredictable war situations. In this paper, we suggest new threat evaluation system based on information fusion JDL model, and it is principle that combine fuzzy which is favorable to refine ambiguous relationships with bayesian network useful to inference battled situation having insufficient evidence and to use learning algorithm. Finally, the system's performance by getting threat evaluation on an air defense scenario is presented.

• Korean Journal of Construction Engineering and Management
• /
• v.23 no.1
• /
• pp.83-88
• /
• 2022

Companies make decisions with risks such as choosing an investment plan in order to pursue profits. This study explained the decision making of the management of construction companies in South Korea using the tendency to avoid losses in the Prospect Theory. To this end, 20-year financial data of 2,881 companies engaged in real estate development, which have to bear the greatest risk among the construction industry, were collected. The collected companies were roughly classified based on the reference point, and the causal relationship between average return on equity and risk preference by group was empirically analyzed through regression analysis. As a result, it was confirmed that if the average return on equity of a company decreases for the group above the reference point, it tends to select an investment plan with low uncertainty in order not to lose additional money. In addition, it was confirmed that if the average return on equity of a company decreases for the group below the reference point, it tends to select an investment plan with high uncertainty to move to the profit area. This result is exactly consistent with the loss aversion tendency of the Prospect Theory.

• Korean Journal of Remote Sensing
• /
• v.33 no.5_3
• /
• pp.799-807
• /
• 2017

Environmental impact assessment is a regulation that is implemented to reduce harmful environmental impacts of development projects. However, The environmental value is difficult to quantify and the uncertainty, There is a problem of objectivity and reliability of the system is consistently pointed out. Therefore, the necessity of the data-based environmental impact assessment system is gradually increasing. Especially, environmental impact assessment is highly applicable to environmental spatial information because it contains about the development of a particular area. Also with the introduction of EIA Follow-up management system, there is a demand for a system for providing and utilizing environmental information in a time series. This study derives the necessity of information provision system and analyz existing environmental information utilization system based on the institutional characteristics of environmental impact assessment. And suggest ways to provide environmental spatial information to find policy implications for improving the limit of environmental impact assessment system.