• Korean Journal of Agricultural and Forest Meteorology
• /
• v.25 no.2
• /
• pp.80-98
• /
• 2023

This study explores the effectiveness of various data preprocessing algorithms for improving subseasonal to seasonal (S2S) climate predictions from six climate forecast models and their Multi-Model Ensemble (MME) using a deep learning-based postprocessing model. A pipeline of data transformation algorithms was constructed to convert raw S2S prediction data into the training data processed with several statistical distribution. A dimensionality reduction algorithm for selecting features through rankings of correlation coefficients between the observed and the input data. The training model in the study was designed with TimeDistributed wrapper applied to all convolutional layers of U-Net: The TimeDistributed wrapper allows a U-Net convolutional layer to be directly applied to 5-dimensional time series data while maintaining the time axis of data, but every input should be at least 3D in U-Net. We found that Robust and Standard transformation algorithms are most suitable for improving S2S predictions. The dimensionality reduction based on feature selections did not significantly improve predictions of daily precipitation for six climate models and even worsened predictions of daily maximum and minimum temperatures. While deep learning-based postprocessing was also improved MME S2S precipitation predictions, it did not have a significant effect on temperature predictions, particularly for the lead time of weeks 1 and 2. Further research is needed to develop an optimal deep learning model for improving S2S temperature predictions by testing various models and parameters.

• Ocean and Polar Research
• /
• v.26 no.4
• /
• pp.581-594
• /
• 2004

CREAHS II carried out an intensive hydrographic survey covering almost entire East Sea in 1999. Hydrographic data from total 203 stations were released to public on the internee. This paper summarized the results of water mass analysis by OHP (Optimum Multiparameter) method that utilizes temperature, salinity, dissolved oxygen, pH, alkalinity, silicate, nitrate, phosphate and location data as an input data-matrix. A total of eight source water types are identified in the East Sea: four in surface waters(North Korea Surface Water, Tatar Surface Cold Water, East Korean Coastal Water, Modified Tsushima Surface Water), two intermediate water types (Tsushima Middle Water, Liman Cold Water), two deep water types (East Sea Intermediate Water, East Sea Proper Water). Of these NKSW, MTSW and TSCW are the newly reported as the source water type. Distribution of each water types reveals several few interesting hydrographic features. A few noteworthy are summarized as follows: The Tsushima Warm Current enter the East Sea as three branches; East Korea Coastal Water propagates north along the coast around $38^{\circ}N$ then turns to northeastward to $42^{\circ}N$ and moves eastward. Cold waters of northern origin move southward along the coast at the subsurface, which existence the existence of a circulation cell at the intermediate depth of the East Sea. The estimated volume of each water types inferred from the OMP results show that the deep waters (ESIW + ESPW) fill up ca. 90% of the East Sea basins. Consequently the formation and circulation of deep waters are the key factors controlling environmental condition of the East Sea.

• Journal of the Korean earth science society
• /
• v.21 no.4
• /
• pp.449-468
• /
• 2000

The Ordovician Chongson Limestone deposited in the carbonate ramp to the rimmed shelf shows diverse diagenetic features. The marine diagenetic feature appears as isopachous cements surrounding ooids and peloids. Meteoric diagenetic features are recrystallized finely and coarsely crystalline calcite, evaporite casts filled with calcite, and isopachous sparry calcite surrounding ooid grains. Shallow burial diagenetic features include wispy seam, microstylolite, and dissolution seam whereas deep burial features include stylolite, burial cements. blocky calcite with twin lamellae, and poikilotopic calcite. Dolomites consist of very finely to finely crystalline mosaic dolomite formed as supratidal dolomite, disseminated dolomite of diverse origin, patchy dolomite formed from bioturbated mottles, and saddle dolomite of burial origin. Silicified features include calcite-replacing quartz and fracture-filling megaquartz. Burial cements characterized by poikilotopic texture show ${\delta}^{18}$O value of -10.4 %$_o$ PDB, ${\delta}^{13}$C value of -1.0%$_o$ PDB and 504ppm Sr, 3643ppm Fe, and 152ppm Mn concentrations. Finely and coarsely crystalline limestones show similar ${\delta}^{18}$O and ${\delta}^{13}$C value to those of burial cements; however, they show lower Sr and higher Fe and Mn concentrations than burial cements. This suggests that very finely and coarsely crystalline limestones were recrystallized in freshwater and then they were readjusted geochemically in the burial setting whereas the burial cements were formed in relatively high temperature and low water/rock ratio conditions. Very finely and finely crystalline mosaic dolomites with ${\delta}^{18}$O value of -8.2%$_o$ PDB, ${\delta}^{13}$C value of -1.9 %$_o$ PDB, and 213ppm Sr, 3654ppm Fe, and 114ppm Mn concentrations, respectively are interpreted to have been formed penecontemporaneously in supratidal flat and then recrystallized in the low water/rock ratio burial environment. Geochemical data suggest that the low water/rock ratio burial environment was the dominant diagenetic setting in the Chongson Limestone. The Chongson Limestone has experienced marine and meteoric diagenesis during early diagenesis. With deposition of Haengmae and Hoedongri formations part of the Chongson Limestone was buried beneath these formations and it experienced shallow burial diagenesis. During the Devonian the Chongson Limestone was tectonically deformed and subaerially exposed. During the Carboniferous to the Permian about 3.3km thick Pyongan Supergroup was deposited on the Chongson Limestone and the Chongson Limestone was in deep burial depths and stylolite, burial cements, blocky calcite and saddle dolomite were formed. After this burial event the Chongson Limestone was subaerially exposed during the Mesozoic and Cenozoic by three periods of tectonic disturbance including Songnim, Daebo and Bulguksa disturbance. Since the Bulguksa disturbance during Cretaceous and early Tertiary the Chongson Limestone has been subaerially exposed.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• Journal of IKEEE
• /
• v.26 no.3
• /
• pp.408-415
• /
• 2022

In this paper, we propose a 3D point cloud reconstruction technique from 2D images using efficient feature map extraction network. The originality of the method proposed in this paper is as follows. First, we use a new feature map extraction network that is about 27% efficient than existing techniques in terms of memory. The proposed network does not reduce the size to the middle of the deep learning network, so important information required for 3D point cloud reconstruction is not lost. We solved the memory increase problem caused by the non-reduced image size by reducing the number of channels and by efficiently configuring the deep learning network to be shallow. Second, by preserving the high-resolution features of the 2D image, the accuracy can be further improved than that of the conventional technique. The feature map extracted from the non-reduced image contains more detailed information than the existing method, which can further improve the reconstruction accuracy of the 3D point cloud. Third, we use a divergence loss that does not require shooting information. The fact that not only the 2D image but also the shooting angle is required for learning, the dataset must contain detailed information and it is a disadvantage that makes it difficult to construct the dataset. In this paper, the accuracy of the reconstruction of the 3D point cloud can be increased by increasing the diversity of information through randomness without additional shooting information. In order to objectively evaluate the performance of the proposed method, using the ShapeNet dataset and using the same method as in the comparative papers, the CD value of the method proposed in this paper is 5.87, the EMD value is 5.81, and the FLOPs value is 2.9G. It was calculated. On the other hand, the lower the CD and EMD values, the better the accuracy of the reconstructed 3D point cloud approaches the original. In addition, the lower the number of FLOPs, the less memory is required for the deep learning network. Therefore, the CD, EMD, and FLOPs performance evaluation results of the proposed method showed about 27% improvement in memory and 6.3% in terms of accuracy compared to the methods in other papers, demonstrating objective performance.

• Journal of the Korean Society of Radiology
• /
• v.18 no.3
• /
• pp.187-201
• /
• 2024

Osteoporosis is a major health issue globally, often remaining undetected until a fracture occurs. To facilitate early detection, deep learning (DL) models were developed to classify osteoporosis using abdominal computed tomography (CT) scans. This study was conducted using retrospectively collected data from 3,012 contrast-enhanced abdominal CT scans. The DL models developed in this study were constructed for using image data, demographic/clinical information, and multi-modality data, respectively. Patients were categorized into the normal, osteopenia, and osteoporosis groups based on their T-scores, obtained from dual-energy X-ray absorptiometry, into normal, osteopenia, and osteoporosis groups. The models showed high accuracy and effectiveness, with the combined data model performing the best, achieving an area under the receiver operating characteristic curve of 0.94 and an accuracy of 0.80. The image-based model also performed well, while the demographic data model had lower accuracy and effectiveness. In addition, the DL model was interpreted by gradient-weighted class activation mapping (Grad-CAM) to highlight clinically relevant features in the images, revealing the femoral neck as a common site for fractures. The study shows that DL can accurately identify osteoporosis stages from clinical data, indicating the potential of abdominal CT scans in early osteoporosis detection and reducing fracture risks with prompt treatment.

• Geophysics and Geophysical Exploration
• /
• v.16 no.3
• /
• pp.163-170
• /
• 2013

The amplitude variation with offset of seismic data can detect fluids in the sediment and resolve the petrophysical properties of hydrocarbons in the subsurface. We analyzed and described the amplitude variation in deep sea seismic data obtained from the Ulleung Basin, East Sea. By inspecting seismic CDP-offset and CDP-angle gathers which show a bright reflection event, we decided a target zone for amplitude variation analysis. From the seismic angle gather at the middle of Ulleung Basin, we recognized amplitude increase or decrease versus offset on the intercept-gradient curve. Using the product attribute and Poisson's ratio change attribute computed in terms of intercept with gradient, the top and the base of gas saturated sediments were described. The area of amplitude variation suggestive of the presence of gas saturated sediments is shown at the depth of 3 s traveltime. Anomalous features of seismic amplitude in the Ulleung Basin were classified by the crossplot of intercept and gradient. The background trend of crossplot between intercept and gradient shows an inverse proportional relation that is common for wet sediments. Anomalous amplitudes of Class III fall into the first and the third quadrants on crossplots. We inferred regional gas/water saturated area with the horizontal dimension of 150 m in the Ulleung Basin by cross-section with respect to cross-plot anomaly.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.15 no.1
• /
• pp.63-79
• /
• 1993

To repair bony defects with tansplanted bone in the body, fresh autogenous bone is undoubtly, the most effective bone graft for clinical applications. But the demineralized bone has the matrix-induced bone formation which was suggested by Urist in 1965. Many authors assisted that demineralized bone powder induces phenotypic conversion of mesenchymal cells into osteoblasts, with high-density bone formation. The process of inducing differentiated cells becomes osteogenic properties. The purpose of this study was to evaluate the osteoinductive capacity of allogenic freeze-dried demineralized bone block (FDD, $7{\times}7mm$) and to compare FDD with the same sue of deep-frozen allogenic bone(DF), fresh autogenous bone (A) after implantation. The histological and ultrastructural features of tissue responses were examined after 1, 2, 4, 6, 8 weeks implantation of each experimental groups in the operative site of the New Zealand white rabbits. The results were as follows : 1. Inflammatory cell infiltration generally has appeared at 1 week, but reduced at 4 weeks in each group, but most severe in DF group. 2. Osteoblastic activity has increased for 4 weeks, but decreased at 6 weeks in each group and there was no significant difference among experimental groups. 3. New bone formation has begun at 1week, least activations in A groups, and showed the revesal line of bone formation among each group at 6 to 8 weeks. 4. Bone resorption has appeared at 1 week, but disappeared at 4 weeks in both A and DF groups, but more severe in DF than A groups. 5. In ultrastructural changs, the DF group have showed the most remarkable osteoclastic activities among experimental groups. 6. Osteoid or tangled collagen fibrils near the implanted sites were replaced by more mature, lamellated bony trabeculae during bone remodeling. There was little difference among each experimental groups. 7. During the convertion osteoblasts to osteocytes which embedded within the bone matrix, there was organ-less-poor cytoplasm, increased nuclear chromatin, abundant rough endothelial reticulum (RER) in each groups. From the above the findings, the DF group shored more bone resorption and foreign body reaction than FDD and A groups, and FDD group showed more new bone formation or osteoblastic activity than DF and A groups in early stage. There was no significant difference of cellular activities among the FDD DF, and A groups according to the time.

• Journal of the Korean earth science society
• /
• v.34 no.3
• /
• pp.195-208
• /
• 2013

We employed electrical resistivity and optical borehole imaging methods to identify underground cavities and determine ground subsidence rate at the study area affected by land subsidence due to abandoned underground mines. At the study site 1, the anomalous zones of low resistivity ranging between 100 ohm-meter and 150 ohm-meter were observed and confirmed as an abandoned underground mine by subsequent borehole drilling and optical borehole imaging. Although the electrical resistivity survey was unavailable due to the paved surface of the study site 2, we were able to locate another abandoned underground mine with the collapsed mine shaft based on the distribution of the ore veins and confirmed it with borehole drilling. In addition, we measured vertical displacements of underground features indicating underground subsidence by conducting optical borehole imaging 6 times over a period of 43 days at the study site 2. The displacement magnitude at the deep segment caused by subsidence appeared to be 3 times larger than those at the shallow segment. Similarly, the displacement duration at the deep segment was 4 times longer than those at the shallow segment. Therefore, the combination of electrical resistivity and optical borehole imaging methods can be effectively applicable to detect and monitor ground subsidence caused by underground cavities.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.7 no.4
• /
• pp.191-205
• /
• 2009

To characterize the geological features in the study area for high-level radioactive waste disposal research, KAERI (Korea Atomic Energy Research Institute) has been performing several geological investigations such as geophysical surveys and borehole drillings since 1997. Especially, the KURT (KAERI Underground Research Tunnel) constructed to understand the deep geological environments in 2006. Recently, the deep boreholes, which have 500 m depth inside the left research module of the KURT and 1,000 m depth outside the KURT, were drilled to confirm and validate the results from a geological model. The objective of this research was to investigate hydrogeological conditions using a 3-D geological model around the KURT. The geological analysis from the surface and borehole investigations determined four important geologicla elements including subsurface weathered zone, low-angled fractures zone, fracture zones and bedrock for the geological model. In addition, the geometries of these elements were also calculated for the three-dimensional model. The results from 3-D geological model in this study will be beneficial to understand hydrogeological environment in the study area as an important part of high-level radioactive waste disposal technology.