• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.309-327
• /
• 2021

Dynamic treatment regimes (DTRs) are decision-making rules designed to provide personalized treatment to individuals in multi-stage randomized trials. Unlike classical methods, in which all individuals are prescribed the same type of treatment, DTRs prescribe patient-tailored treatments which take into account individual characteristics that may change over time. The Q-learning method, one of regression-based algorithms to figure out optimal treatment rules, becomes more popular as it can be easily implemented. However, the performance of the Q-learning algorithm heavily relies on the correct specification of the Q-function for response, especially in observational studies. In this article, we examine a number of double-robust weighted least-squares estimating methods for Q-learning in high-dimensional settings, where treatment models for propensity score and penalization for sparse estimation are also investigated. We further consider flexible ensemble machine learning methods for the treatment model to achieve double-robustness, so that optimal decision rule can be correctly estimated as long as at least one of the outcome model or treatment model is correct. Extensive simulation studies show that the proposed methods work well with practical sample sizes. The practical utility of the proposed methods is proven with real data example.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.121-133
• /
• 2020

This study was carried out to set the direction of the new industry policy of Busan city by analyzing the changing trend of artificial intelligence technology that has recently developed rapidly and predicting the direction of future development. The company wanted to draw up support measures to utilize artificial intelligence technology, which has been rapidly emerging in the market, in the region's specialized industry. Artificial intelligence is a key keyword in the fourth industrial revolution and artificial intelligence-based data utilization technology can be used in various fields from manufacturing processes to services, and is entering an era of super-fusion in which barriers between technologies and industries will be broken down. In this study, the direction of promotion for fostering Busan as an artificial intelligence city was derived based on the comparison and analysis of artificial intelligence-related ecosystems among major local governments. In this study, we wanted to present a plan to create an artificial intelligence industrial ecosystem that can be called a key policy to foster Busan as an 'AI City'. Busan's plan to foster the AI industry ecosystem is aimed at establishing a policy direction to ultimately nurture the artificial intelligence industry as Busan's future food source.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.157-165
• /
• 2020

Aiming at the development of neural network and self-driving data set, it is also an idea to improve the performance of network model to detect moving objects by dividing the data set. In Darknet network framework, the YOLOv4 (You Only Look Once v4) network model was used to train and test Udacity data set. According to 7 proportions of the Udacity data set, it was divided into three subsets including training set, validation set and test set. K-means++ algorithm was used to conduct dimensional clustering of object boxes in 7 groups. By adjusting the super parameters of YOLOv4 network for training, Optimal model parameters for 7 groups were obtained respectively. These model parameters were used to detect and compare 7 test sets respectively. The experimental results showed that YOLOv4 can effectively detect the large, medium and small moving objects represented by Truck, Car and Pedestrian in the Udacity data set. When the ratio of training set, validation set and test set is 7:1.5:1.5, the optimal model parameters of the YOLOv4 have highest detection performance. The values show mAP50 reaching 80.89%, mAP75 reaching 47.08%, and the detection speed reaching 10.56 FPS.

• Atmosphere
• /
• v.31 no.1
• /
• pp.113-141
• /
• 2021

The impacts of ice clouds on the energy budget of the Earth and their representation in climate models have been identified as important and unsolved problems. Ice clouds consist almost exclusively of non-spherical ice crystals with various shapes and sizes. To determine the influences of ice clouds on solar and infrared radiation as required for remote sensing retrievals and numerical models, knowledge of scattering and microphysical properties of ice crystals is required. A conventional method for representing the radiative properties of ice clouds in satellite retrieval algorithms and numerical models is to combine measured microphysical properties of ice crystals from field campaigns and pre-calculated single-scattering libraries of different shapes and sizes of ice crystals, which depend heavily on microphysical and scattering properties of ice crystals. However, large discrepancies between theoretical calculations and observations of the radiative properties of ice clouds have been reported. Electron microscopy images of ice crystals grown in laboratories and captured by balloons show varying degrees of complex morphologies in sub-micron (e.g., surface roughness) and super-micron (e.g., inhomogeneous internal and external structures) scales that may cause these discrepancies. In this study, the current idealized models representing morphologies of ice crystals and the corresponding numerical methods (e.g., geometric optics, discrete dipole approximation, T-matrix, etc.) to calculate the single-scattering properties of ice crystals are reviewed. Current problems and difficulties in the calculations of the single-scattering properties of atmospheric ice crystals are addressed in terms of cloud microphysics. Future directions to develop physically consistent ice-crystal models are also discussed.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.24 no.6
• /
• pp.129-137
• /
• 2020

In performing the structural analysis, the foundation is considered to be a fixed end as a plastic hinge model. In this study, the displacements of the foundation, pier, and shoe were compared when the foundation modeled as a fixed end, a shallow foundation constructed on bedrock of 2m depth, and a pile foundation constructed in the 10m to 20m depth of bedrock. The shear force was also compared, and the probability of damage was calculated and compared for the critical condition. When calculated as a fixed end, the displacement of the foundation converged to 0mm, but the shallow foundation built on the bedrock with a depth of 2m caused relatively displacement, and the pile foundation constructed to contact the bedrock with a depth of 18m caused a larger displacement. In addition, it was analyzed that the displacement of the foundation, which is the lower structure, affects the displacement of the super structure, but the difference in shear force applied to the foundation was insignificant in the three cases. There was no difference between the shallow foundation and the pile foundation in the influence on the displacement of the top of the pier, but there was a big difference from the analysis assuming as a fixed end.

• Journal of Space Technology and Applications
• /
• v.2 no.2
• /
• pp.146-168
• /
• 2022

Witnessing current military conflicts in South China Sea and Eastern Europe, most defense analysts evaluate one of the most serious security threat toward the US is coming from the superpower competitions with Russia and China. The main means for such super power hegemonic competitions is military power and space power is a key enabler to maximize the efficiency and effectiveness of military employment. Reflecting above circumstances, the space hegemonic competition between the Unites States and China is spreading into all aspects of national powers. Under such an environment, R.O.K needs to significantly develop national space power to preserve life and assets of people in space. On the other hand, the R.O.K has a lot of limitations in launching space assets into orbits by land-based space rockets due to its geographic locations. The limitation of rocket launching direction, the failure to secure a significant area enough to secure safety and the limitation to secure open area enough to build associated facilities are among them. On this paper, I will suggest the need to build the 2nd space rocket launching site after analyzing a lot of short-falls the current 'Naro' space center face, compared to those of advanced space powers around the world.

• Journal of Korea Foundry Society
• /
• v.41 no.6
• /
• pp.505-515
• /
• 2021

Creep properties of directionally solidified Ni-based superalloy CM247LC under various temperature and stress conditions have been investigated. In the heat-treated specimen, some portion of eutectic γ-γ' remained, and uniform cubic γ' was observed in the dendrites. At low temperature (750℃) and high stress condition, a large amount of deformation occurred during the primary creep, while the tertiary creep region accounted for most of the creep deformation under high temperature and low stress condition. γ' particles are sheared by dislocation dissociated into super lattice partial dislocations separated by stacking faults at 750℃. No stacking faults in γ' were found at and above 850℃ due to the temperature dependence of the stacking fault energy. Raft structure of γ' was found after creep test at high temperature of 950℃ and 1000℃. At 850℃, the deformation mechanism was shown to be dependent on the stress condition, and so rafting was observed only under low stress condition.

• Korean Journal of Remote Sensing
• /
• v.38 no.2
• /
• pp.167-177
• /
• 2022

In order to obtain satellite image products using the image transmitted to the ground station after capturing the satellite images, many image pre/post-processing steps are involved. During the pre/post-processing, when converting from level 1R images to level 1G images, geometric correction is essential. An interpolation method necessary for geometric correction is inevitably used, and the quality of the level 1G images is determined according to the accuracy of the interpolation method. Also, it is crucial to speed up the interpolation algorithm by the level processor. In this paper, we proposed a lightweight CNN-based interpolation method required for geometric correction when converting from level 1R to level 1G. The proposed method doubles the resolution of satellite images and constructs a deep learning network with a lightweight deep convolutional neural network for fast processing speed. In addition, a feature map fusion method capable of improving the image quality of multispectral (MS) bands using panchromatic (PAN) band information was proposed. The images obtained through the proposed interpolation method improved by about 0.4 dB for the PAN image and about 4.9 dB for the MS image in the quantitative peak signal-to-noise ratio (PSNR) index compared to the existing deep learning-based interpolation methods. In addition, it was confirmed that the time required to acquire an image that is twice the resolution of the 36,500×36,500 input image based on the PAN image size is improved by about 1.6 times compared to the existing deep learning-based interpolation method.

• Molecules and Cells
• /
• v.45 no.7
• /
• pp.465-478
• /
• 2022

MicroRNAs (miRNAs) are a class of small non-coding RNAs that regulate the expression of target messenger RNA (mRNA) complementary to the 3' untranslated region (UTR) at the post-transcriptional level. Hsa-miR-422a, which is commonly known as miRNA derived from transposable element (MDTE), was derived from short interspersed nuclear element (SINE). Through expression analysis, hsa-miR-422a was found to be highly expressed in both the small intestine and liver of crab-eating monkey. AT-Rich Interaction Domain 5 B (ARID5B) was selected as the target gene of hsa-miR-422a, which has two binding sites in both the exon and 3'UTR of ARID5B. To identify the interaction between hsa-miR-422a and ARID5B, a dual luciferase assay was conducted in HepG2 cell line. The luciferase activity of cells treated with the hsa-miR-422a mimic was upregulated and inversely downregulated when both the hsa-miR-422a mimic and inhibitor were administered. Nuclear factor erythroid-2 (NF-E2) was selected as the core transcription factor (TF) via feed forward loop analysis. The luciferase expression was downregulated when both the hsa-miR-422a mimic and siRNA of NF-E2 were treated, compared to the treatment of the hsa-miR-422a mimic alone. The present study suggests that hsa-miR-422a derived from SINE could bind to the exon region as well as the 3'UTR of ARID5B. Additionally, hsa-miR-422a was found to share binding sites in ARID5B with several TFs, including NF-E2. The hsa-miR-422a might thus interact with TF to regulate the expression of ARID5B, as demonstrated experimentally. Altogether, hsa-miR-422a acts as a super enhancer miRNA of ARID5B by collaborating with TF and NF-E2.

• Journal of Korea Port Economic Association
• /
• v.38 no.3
• /
• pp.1-13
• /
• 2022

The Busan-Ulsan-Gyeongsangmando(BuUlGyeong) megacity aims to grow into the second metropolitan area by strengthening the growth and connectivity of the southeast region in Korea. In order to analyze the linkage between port logistics and regional economic development, this study focused on the economic community of BuUlGyeong. The findings are as follows. First, it was found that vessel entry and departure had a positive (+) significant effect on cargo volume. Second, the cargo volume was found to have a preemptive effect on the vessel entry and exit ports and the composite index. Third, the composite Index was found to have a significant positive (+) effect on cargo volume. From the results of the variance decomposition, it can be confirmed that the composite index has a weak impact on ship entry/exit and cargo volume. Based on the above findings, some policy implications are derived. First, regional governments in BuUlGyeong should provide incentives to revitalize the maritime economy such as the expansion of maritime port infrastructure and encouraging investment. Second, since an increase in cargo transport volume at major national ports linked to the BuUlGyeong region has a positive (+) effect on economy. Therefore it is necessary to expand actively port marketing and regular service for the country and ports. Third, investments should be made continuously to expand the marine port infrastructure such as the expansion of facilities so that global shipping companies can continue to call megacities in BuUlGyeong ports for super-large ships and cruises.