http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
통일의 상상력과 시적 글쓰기의 지평 ― 조태일의 시와 시론을 중심으로 ―
정민구 ( Jeong¸ Min-gu ) 현대문학이론학회 2020 現代文學理論硏究 Vol.0 No.83
2018년 4월 27일, 판문점 회의실과 중립국감독위원회 캠프 사이에 놓인 도보다리 위에서 남북 정상이 대화를 나눈 이후로, 미지의 영역에 놓여있던 통일은 일련의 정세를 형성하면서 현실의 영역으로 다가오게 되었다. 도래하는 통일의 정세와 마주하면서, 이 글에서는 오늘날의 문학, 특히 시가 실천할 수 있는 통일에의 과업은 무엇인가에 대해 모색해보고자 하였다. 물음에 대한 단초를 마련하기 위해, 이 글은 조태일의 시적 글쓰기에 나타난 통일의 상상력에 주목하였다. 이때의 시적 글쓰기는 시인이 창조한 시와 시론을 포함하는 것이다. 아울러 통일의 상상력이란 통일을 지향하는 시적 글쓰기의 토대가 되는 창조적 상상력을 말하는 것이다. 실천적 글쓰기라는 관점에서 조태일의 시적 글쓰기는 통일을 향한 창조적 상상력에 기반하여 통일국토라는 경험공간의 역사성에 대한 진솔한 깨달음과 미래에 대한 긍정적인 전망을 포월하고 있다. 이는 오늘날 희미해지고 있는 통일에의 과업을 다시 환기시켜 준다는 차원에서뿐만 아니라 새롭게 부상하고 있는 매체와의 결합 가능성도 담지하고 있다는 측면에서, 도래하는 통일시대의 목전에서 시적 글쓰기의 지평을 전망할 수 있었다. On April 27, 2018, the leaders of the two Koreas talked on a footbridge between the Panmunjom conference room and the Neutral Nations Supervisory Commission camp. Since then, unification, which had been in uncharted territory, has come as a realm of reality, forming a kind of political situation. Faced with this looming situation of unification, this article sought to explore the tasks of unification that today's literature, especially poetry, could carry out. To seek answers to the question, this article noted the imagination of unification shown in Cho Tae-il's poetic writing. Of course, poetic writing at this time involves poems and essays on poetry. In addition, the imagination of unification refers to the creative imagination that serves as the basis for poetic writing aimed at unification. From the perspective of practical writing, Cho Tae-il's poetic writing, based on his creative imagination toward unification, offers a sincere realization of the historicality of the experience space of unification and a positive outlook for the future. This not only serves as a reminder of the task of unification that is fading today, but also contains the possibility of combining with newly emerging media, which we expected to see the horizon of poetic writing in the coming unification era.
Mingu Kang(강민구),Hyun Joo Kim(김현주),Aera Jang(장애라),Dong Keun Gam(감동근),Gwan Sik Yun(윤관식),Cheorun Jo(조철훈) 충남대학교 농업과학연구소 2012 농업과학연구 Vol.39 No.1
This study was carried out to investigate the effects of dietary supplementation of quercetin (KocetinTM, QR) on antioxidative activity and meat quality of beef cattle (Holstein-Friesian). Beef cattle were divided into 3 groups; dietary supplementation of QR at 21 (n=4) and 42 ppm (n=3), and non-supplemented control (n=4). The QR comprised of 10% of quercetin. After slaughtering the beef cattle, loins were obtained and analyzed. Dietary supplementation of QR at 42 ppm showed significantly higher final pH of loin but did not affect the water holding capacity, drip loss, cooking loss, surface color, total phenolics content, 1,1-diphenyl-2-picrylhydrazyl radical scavenging activity. Dietary QR showed no difference in both 2-thiobarbituric acid reactive substances and volatile basic nitrogen values. Textural characteristic results also showed no difference, except for cohesiveness. Cohesiveness was significantly higher in loin from beef cattle treated by dietary QR at 42 ppm when compared to control. Results suggest that dietary QR, which has only 10% of quercetin is not sufficient to have positive biochemical effects on beef meat quality.
( Mingu Kang ),( Hyeungkyeom Kim ),( Suchul Lee ),( Seokmin Han ) 한국인터넷정보학회 2021 KSII Transactions on Internet and Information Syst Vol.15 No.11
Recently, malware classification based on Deep Neural Networks (DNN) has gained significant attention due to the rise in popularity of artificial intelligence (AI). DNN-based malware classifiers are a novel solution to combat never-before-seen malware families because this approach is able to classify malwares based on structural characteristics rather than requiring particular signatures like traditional malware classifiers. However, these DNN-based classifiers have been found to lack robustness against malwares that are carefully crafted to evade detection. These specially crafted pieces of malware are referred to as adversarial examples. We consider a clever adversary who has a thorough knowledge of DNN-based malware classifiers and will exploit it to generate a crafty malware to fool DNN-based classifiers. In this paper, we propose a DNN-based malware classifier that becomes resilient to these kinds of attacks by exploiting Generative Adversarial Network (GAN) based data augmentation. The experimental results show that the proposed scheme classifies malware, including AEs, with a false positive rate (FPR) of 3.0% and a balanced accuracy of 70.16%. These are respective 26.1% and 18.5% enhancements when compared to a traditional DNN-based classifier that does not exploit GAN.
Mingu Kim,Youdan Kim 한국항공우주학회 2013 International Journal of Aeronautical and Space Sc Vol.14 No.1
Nonlinear guidance law combined with a pseudo pursuit guidance is proposed, to perform stationary target observation mission. Multiple UAVs are considered, with waypoint constraint. The whole guidance is divided into two steps: firstly, waypoint approach, with specified incidence angle; and secondly, loitering around the stationary target. Geometric approach is used to consider the constraint on the waypoint, and a specified phase angle between the loitering UAV and the approaching UAV. In the waypoint approach step, UAVs fly to the waypoint using the pseudo pursuit guidance law. After passing the waypoint, UAVs turn around the target, using a distance error dynamics-based guidance law. Numerical simulations are performed, to verify the performance of the proposed guidance law.
Real-time selective gas detection by gas sensor array and deep learning
Mingu Kang(강민구),Incheol Cho(조인철),Inkyu Park(박인규) 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.11
The demand for gas sensors is increasing because of the growing interest in monitoring indoor/outdoor air pollutions. In particular, semiconductor metal oxide (SMO) gas sensors are attracting attention as the next-generation gas sensors. However, there are limitations in the actual applications of SMO gas sensors due to their low selectivity. In this study, the selectivity problem could be solved by fabricating a gas sensor array and using the deep learning network. The fabricated gas sensor array used nanocolumnar films of metal oxides (SnO₂, In₂O₃, WO₃, and CuO) deposited through the glancing angle deposition (GLAD) as the sensing materials, and the convolutional neural network (CNN) was selected as the deep learning network for gas identification. Finally, a real-time selective gas detection for CO, NH₃, NO₂, Methane, and Acetone gas was achieved with an accuracy of 98% by applying preprocessed sensing data collected from the gas sensor arrays to the CNN.
Mingu Kang,Cheol-Hong Lim,Jeong-Hee Han 한국독성학회 2013 Toxicological Research Vol.29 No.2
Nanotoxicological research has shown toxicity of nanomaterials to be inversely related to particle size. However, the contribution of agglomeration to the toxicity of nanomaterials has not been sufficiently studied, although it is known that agglomeration is associated with increased nanomaterial size. In this study, we prepared aerosols of nano-sized carbon black by 2 different ways to verify the effects of agglomeration on the toxicity and deposition of nano-sized carbon black. The 2 methods of preparation included the carbon black dispersion method that facilitated clustering without sonication and the carbon black dispersion method involving sonication to achieve scattering and deagglomeration. Male Sprague-Dawley rats were exposed to carbon black aerosols 6 hr a day for 3 days or for 2 weeks. The median mass aerodynamic diameter of carbon black aerosols averaged 2.08 μm (for aerosol prepared without sonication; group N) and 1.79 μm (for aerosol prepared without sonication; group S). The average concentration of carbon black during the exposure period for group N and group S was 13.08 ± 3.18 mg/m³ and 13.67 ± 3.54 mg/m³, respectively, in the 3-day experiment. The average concentration during the 2-week experiment was 9.83 ± 3.42 mg/m³ and 9.08 ± 4.49 mg/m³ for group N and group S, respectively. The amount of carbon black deposition in the lungs was significantly higher in group S than in group N in both 3-day and 2-week experiments. The number of total cells, macrophages and polymorphonuclear leukocytes in the bronchoalveolar lavage (BAL) fluid, and the number of total white blood cells and neutrophils in the blood in the 2-week experiment were significantly higher in group S than in normal control. However, differences were not found in the inflammatory cytokine levels (IL-1β, TNF-α, IL-6, etc.) and protein indicators of cell damage (albumin and lactate dehydrogenase) in the BAL fluid of both group N and group S as compared to the normal control. In conclusion, carbon black aerosol generated by sonication possesses smaller nanoparticles that are deposited to a greater extent in the lungs than is aerosol formulated without sonication. Additionally, rats were narrowly more affected when exposed to carbon black aerosol generated by sonication as compared to that produced without sonication.