• 한국지능시스템학회논문지
• /
• 제19권4호
• /
• pp.568-573
• /
• 2009

해양사고로 인한 경제적, 환경적 피해가 커짐에 따라, 해양사고 방지를 위한 이슈가 크게 대두되고 있다. 발생된 해양사고 사례의 종류와 원인을 분석하여 구축된 DB가 해양사고 방지를 위한 연구에 널리 활용 되고 있으나, 하나의 종류 및 원인에 대해서만 DB가 구축되어 있어 일반적으로 복수의 원인에 의해 발생되고 복수의 종류에 해당하는 해양사고를 합리적으로 분류하지 못하고 다양하고 막연한 조건을 이용해 검색할 수 없다는 문제점이 있다. 따라서 본 연구에서는 FCM을 이용하여 복수의 해양사고 원인과 종류에 연계된 해양사고 DB를 구축하고 언어 레이블을 이용하여 다양한 원인과 종류에 의해 해양사고 사례추출이 가능한 검색 시스템을 제시하였다.

• Journal of Microbiology and Biotechnology
• /
• 제27권1호
• /
• pp.72-76
• /
• 2017

Detection of exon 19 deletion mutation in the epidermal growth factor receptor (EGFR) gene, which results in increased and sustained phosphorylation of EGFR, is important for diagnosis and treatment guidelines in non-small-cell lung cancer. Here, we have developed a simple and convenient detection system using the interaction between G-quadruplex and fluorophore thioflavin T (ThT) for discriminating EGFR exon 19 deletion mutant DNA from wild type without a label and quencher. In the presence of exon 19 deletion mutant DNA, the probe DNAs annealed to the target sequences were transformed into G-quadruplex structure. Subsequent intercalation of ThT into the G-quadruplex resulted in a light-up fluorescence signal, which reflects the amount of mutant DNA. Due to stark differences in fluorescence intensity between mutant and wild-type DNA, we suggest that the induced G-quadruplex structure in the probe DNA can report the presence of cancer-causing deletion mutant DNAs with high sensitivity.

• 한국주거학회논문집
• /
• 제25권2호
• /
• pp.71-78
• /
• 2014

In recent years, Korean government has focused on improving the environmental impact of products in order to reduce greenhouse gas emissions and to achieve their energy goals. The government has been conducting the following polices such as green procurement inducement and certification system. After carbon labeling was conducted in 2009, among a total of 1,065 items, 97 building materials have been given a certification: finishing materials items have the highest weight (56%). The increase in the certification numbers shows that there has been considerable technical efforts in the building material industry. At the awareness of carbon label and purchase of low carbon product, however, customers are aware of carbon labeling but the purchasing rate of carbon product is low. In this paper, we suggest that low carbon activities must also be considered in order to create client value by adding the concept of ecological efficiency. The objective of this study to measurer the eco-efficiency of carbon labeled building materials on the basis of environmental aspects of the product with the perspective of economy for purchasing the excellent products.

• BMB Reports
• /
• 제41권2호
• /
• pp.126-131
• /
• 2008

The folding of aptamer immobilized on an Au electrode was successfully detected using label-free electrochemical methods. A thrombin binding DNA aptamer was used as a model system in the presence of various monovalent cations. Impedance spectra showed that the extent to which monovalent cations assist in folding of aptamer is ordered as $K^+$ > $NH_4^+$ > $Na^+$ > $Cs^+$. Our XPS analysis also showed that $K^+$ and $NH_4^+$ caused a conformational change of the aptamer in which it forms a stable complex with these monovalent ions. Impedance results for the interaction between aptamer and thrombin indicated that thrombin interacts more with folded aptamer than with unfolded aptamer. The EQCM technique provided a quantitative analysis of these results. In particular, the present impedance results showed that thrombin participates a folding of aptamer to some extent, and XPS analysis confirmed that thrombin stabilizes and induces the folding of aptamer.

• ETRI Journal
• /
• 제27권6호
• /
• pp.733-746
• /
• 2005

In the context of multi-protocol label switching (MPLS) traffic engineering, this paper proposes a scalable constraintbased shortest path first (CSPF) routing algorithm with multiple QoS metrics. This algorithm, called the multiple constraint-based shortest path first (M_CSPF) algorithm, provides an optimal route for setting up a label switched path (LSP) that meets bandwidth and end-to-end delay constraints. In order to maximize the LSP accommodation probability, we propose a link weight computation algorithm to assign the link weight while taking into account the future traffic load and link interference and adopting the concept of a critical link from the minimum interference routing algorithm. In addition, we propose a bounded order assignment algorithm (BOAA) that assigns the appropriate order to the node and link, taking into account the delay constraint and hop count. In particular, BOAA is designed to achieve fast LSP route computation by pruning any portion of the network topology that exceeds the end-to-end delay constraint in the process of traversing the network topology. To clarify the M_CSPF and the existing CSPF routing algorithms, this paper evaluates them from the perspectives of network resource utilization efficiency, end-to-end quality, LSP rejection probability, and LSP route computation performance under various network topologies and conditions.

• 한국인터넷방송통신학회논문지
• /
• 제15권3호
• /
• pp.59-65
• /
• 2015

최근 네트워크 산업계에서 기존 전송 망 보다 유연한 망을 구축하기 위한 캐리어이더넷 기술이 부각되고 있으며 MPLS-TP (MultiProtocol Label Switching-Transport Profile) 기술이 해당 영역의 주요 표준 기술로 적용되고 있다. 하지만 MPLS-TP 망내의 장비를 관리하기 위한 관리 망에 대해서는 명확하지 않은 상태이다. 본 논문에서는 MPLS-TP 망 내의 장비 설치 시 부가적 설정 없이 자동으로 관리 망을 구축하는 방안을 제안한다. 제안하는 방법으로 관리시스템과 망 연결을 위한 설정을 최소화 하여 망 내의 장치 설치비용과 유지 관리 비용을 감축할 수 있다. 또한 실제 MPLS-TP 장비에 적용하여 제안하는 방법의 실효성에 대해 검증하였다.

• 전기학회논문지
• /
• 제67권9호
• /
• pp.1224-1231
• /
• 2018

Automatic age estimation has been used in many social network applications, practical commercial applications, and human-computer interaction visual-surveillance biometrics. However, it has rarely been explored. In this paper, we propose an automatic age estimation system, which includes face detection and convolutional deep learning based on an inception module. The latter is a 22-layer-deep network that serves as the particular category of the inception design. To evaluate the proposed approach, we use 4,000 images of eight different age groups from the Adience age dataset. k-fold cross-validation (k = 5) is applied. A comparison of the performance of the proposed work and recent related methods is presented. The results show that the proposed method significantly outperforms existing methods in terms of the exact accuracy and off-by-one accuracy. The off-by-one accuracy is when the result is off by one adjacent age label to the above or below. For the exact accuracy, the age label of "60+" is classified with the highest accuracy of 76%.

• International journal of advanced smart convergence
• /
• 제8권1호
• /
• pp.141-146
• /
• 2019

These days, there are many applications using neural networks as parts of their system. On the other hand, adversarial examples have become an important issue concerining the security of neural networks. A classifier in neural networks can be fooled and make it miss-classified by adversarial examples. There are many research to encounter adversarial examples by using denoising methods. Some of them using GAN (Generative Adversarial Network) in order to remove adversarial noise from input images. By producing an image from generator network that is close enough to the original clean image, the adversarial examples effects can be reduced. However, there is a chance when adversarial noise can survive the approximation process because it is not like a normal noise. In this chance, we propose a research that utilizes high-level representation in the classifier by combining GAN network with a trained U-Net network. This approach focuses on minimizing the loss function on high representation terms, in order to minimize the difference between the high representation level of the clean data and the approximated output of the noisy data in the training dataset. Furthermore, the generated output is checked whether it shows minimum error compared to true label or not. U-Net network is trained with true label to make sure the generated output gives minimum error in the end. At last, the remaining adversarial noise that still exist after low-level approximation can be removed with the U-Net, because of the minimization on high representation terms.

• Journal of Korean Neurosurgical Society
• /
• 제66권3호
• /
• pp.308-315
• /
• 2023

Recent updates in genomic-integrated glioma classification have caused confusion in current clinical practice, as management protocols and health insurance systems are based on evidence from previous diagnostic classifications. The Korean Brain Tumor Society conducted an electronic questionnaire for society members, asking for their ideas on risk group categorization and preferred treatment for each individual diagnosis listed in the new World Health Organization (WHO) classification of gliomas. Additionally, the current off-label drug use (OLDU) protocols for glioma management approved by the Health Insurance Review and Assessment Service (HIRA) in Korea were investigated. A total of 24 responses were collected from 20 major institutes in Korea. A consensus was reached on the dichotomic definition of risk groups for glioma prognosis, using age, performance status, and extent of resection. In selecting management protocols, there was general consistency in decisions according to the WHO grade and the risk group, regardless of the individual diagnosis. As of December 2022, there were 22 OLDU protocols available for the management of gliomas in Korea. The consensus and available options described in this report will be temporarily helpful until there is an accumulation of evidence for effective management under the new classification system for gliomas.

• Nuclear Engineering and Technology
• /
• 제56권2호
• /
• pp.558-567
• /
• 2024

Diagnostic support systems are being researched to assist operators in identifying and responding to abnormal events in a nuclear power plant. Most studies to date have considered single abnormal events only, for which it is relatively straightforward to obtain data to train the deep learning model of the diagnostic support system. However, cases in which multiple abnormal events occur must also be considered, for which obtaining training data becomes difficult due to the large number of combinations of possible abnormal events. This study proposes an approach to maintain diagnostic performance for multiple abnormal events by training a deep learning model with data on single abnormal events only. The proposed approach is applied to an existing algorithm that can perform feature selection and multi-label classification. We choose an extremely randomized trees classifier to select dedicated monitoring parameters for target abnormal events. In diagnosing each event occurrence independently, two-channel convolutional neural networks are employed as sub-models. The algorithm was tested in a case study with various scenarios, including single and multiple abnormal events. Results demonstrated that the proposed approach maintained diagnostic performance for 15 single abnormal events and significantly improved performance for 105 multiple abnormal events compared to the base model.