• The Journal of Engineering Geology
• /
• v.32 no.4
• /
• pp.697-723
• /
• 2022

Data-driven models to predict groundwater levels 30 days in advance were developed for 12 groundwater monitoring stations in the middle-Jeju watershed, Jeju Island. Stacked long short-term memory (stacked-LSTM), a deep learning technique suitable for time series forecasting, was used for model development. Daily time series data from 2001 to 2022 for precipitation, groundwater usage amount, and groundwater level were considered. Various models were proposed that used different combinations of the input data types and varying lengths of previous time series data for each input variable. A general procedure for deep-learning-based model development is suggested based on consideration of the comparative validation results of the tested models. A model using precipitation, groundwater usage amount, and previous groundwater level data as input variables outperformed any model neglecting one or more of these data categories. Using extended sequences of these past data improved the predictions, possibly owing to the long delay time between precipitation and groundwater recharge, which results from the deep groundwater level in Jeju Island. However, limiting the range of considered groundwater usage data that significantly affected the groundwater level fluctuation (rather than using all the groundwater usage data) improved the performance of the predictive model. The developed models can predict the future groundwater level based on the current amount of precipitation and groundwater use. Therefore, the models provide information on the soundness of the aquifer system, which will help to prepare management plans to maintain appropriate groundwater quantities.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.8
• /
• pp.928-941
• /
• 2016

Energy consumption of today's wired data networks is driven by access networks. Today, green networking has become a issue to reduce energy wastes and $CO_2$ emission by adding energy managing mechanism to wired data networks. However, energy consumption and environmental impacts of wired access networks are largely unknown. In addition, there is a lack of general and quantitative valuation basis of energy use of large-scale access networks and $CO_2$ emissions from them. This paper compared and analyzed limits of existing models estimating energy consumption of access networks and it proposed a model to estimate energy consumption of large-scale access networks by top-down approach. In addition, this work presented models that assess environmental(anti-greening) impacts of access networks using results from our models. The performance evaluation of the proposed models are achieved by comparing with previous models based on existing investigated materials and actual measured values in accordance with real cases.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.32 no.3
• /
• pp.149-156
• /
• 2018

Traditional medicines (TM) in Korea, China, and Japan share most of the theories and therapeutic tools, but there are also differences due to their unique histories and cultures. Here, we aim to identify the differences in the utilization of TM theory between three countries by analyzing herb usage data in terms of the related traditional theories. Herb usage data of each country was collected from "Investigation of Korean medicine use and herbal medicine consumption survey" (Korea), "Analytical report on circulation of key Chinese medicinal materials" (China), and "Survey report on raw material crude drug usage" (Japan). Fifty five herbs with sixty features belonging to five theoretical categories (four properties, five tastes, targeting meridians, treatment strategies, and herbal parts) were selected and analyzed. Weight Sum Model (WSM) and Network-Based Group Features (NBGF) were used to compare the theoretical characteristics of TM between three countries. For the statistical evaluation, we developed and applied Herb Set Enrichment Analysis (HSEA) for WSM and NBGF results. HSEA for WSM results revealed the kidney meridian were targeted more in Korea than Japan, while the spleen meridian were targeted more in Japan than Korea. Herbs with sour taste were used more in Japan than China. HSEA for NBGF results found that NBGF including warm, neutral, sweet, and tonifying features were more dominant in Korea and than Japan, while NBGF including cold, bitter, heat-clearing features were more dominant in Japan than the others. These results suggest that TM in Korea, China, and Japan have unique aspects of practice patterns and theoretical utilization.

• Journal of Korea Multimedia Society
• /
• v.22 no.1
• /
• pp.44-54
• /
• 2019

Recently, image-based object detection has made great progress with the introduction of Convolutional Neural Network (CNN). Many trials such as Region-based CNN, Fast R-CNN, and Faster R-CNN, have been proposed for achieving better performance in object detection. YOLO has showed the best performance under consideration of both accuracy and computational complexity. However, these data-driven detection methods including YOLO have the fundamental problem is that they can not guarantee the good performance without a large number of training database. In this paper, we propose a data sampling method using CycleGAN to solve this problem, which can convert styles while retaining the characteristics of a given input image. We will generate the insufficient data samples for training more robust object detection without efforts of collecting more database. We make extensive experimental results using the day-time and night-time road images and we validate the proposed method can improve the object detection accuracy of the night-time without training night-time object databases, because we converts the day-time training images into the synthesized night-time images and we train the detection model with the real day-time images and the synthesized night-time images.

• Progress in Medical Physics
• /
• v.31 no.3
• /
• pp.111-123
• /
• 2020

Deep learning (DL) is a subset of machine learning and artificial intelligence that has a deep neural network with a structure similar to the human neural system and has been trained using big data. DL narrows the gap between data acquisition and meaningful interpretation without explicit programming. It has so far outperformed most classification and regression methods and can automatically learn data representations for specific tasks. The application areas of DL in radiation oncology include classification, semantic segmentation, object detection, image translation and generation, and image captioning. This article tries to understand what is the potential role of DL and what can be more achieved by utilizing it in radiation oncology. With the advances in DL, various studies contributing to the development of radiation oncology were investigated comprehensively. In this article, the radiation treatment process was divided into six consecutive stages as follows: patient assessment, simulation, target and organs-at-risk segmentation, treatment planning, quality assurance, and beam delivery in terms of workflow. Studies using DL were classified and organized according to each radiation treatment process. State-of-the-art studies were identified, and the clinical utilities of those researches were examined. The DL model could provide faster and more accurate solutions to problems faced by oncologists. While the effect of a data-driven approach on improving the quality of care for cancer patients is evidently clear, implementing these methods will require cultural changes at both the professional and institutional levels. We believe this paper will serve as a guide for both clinicians and medical physicists on issues that need to be addressed in time.

• The Research Journal of the Costume Culture
• /
• v.32 no.2
• /
• pp.247-259
• /
• 2024

This study assessed the effectiveness of brand image communication on consumer perceptions of cruelty-free fashion brands. Brand messaging data were gathered from postings on the official Instagram accounts of three cruelty-free fashion brands and consumer perception data were gathered from Tweets containing keywords related to each brand. Web crawling and natural language processing were performed using Python and sentiment analysis was conducted using the BERT model. By analyzing Instagram content from Stella McCartney, Patagonia, and Freitag from their inception until 2021, this study found these brands all emphasize environmental aspects but with differing focuses: Stella McCartney on ecological conservation, Patagonia on an active outdoor image, and Freitag on upcycled products. Keyword analysis further indicated consumers perceive these brands in line with their brand messaging: Stella McCartney as high-end and eco-friendly, Patagonia as active and environmentally conscious, and Freitag as centered on recycling. Results based on the assessment of the alignment between brand-driven images and consumer-perceived images and the sentiment evaluation of the brand confirmed the outcomes of brand communication performance. The study revealed a correlation between brand image and positive consumer evaluations, indicating that higher alignment of ethical values leads to more positive consumer assessments. Given that consumers tend to prioritize search keywords over brand concepts, it's important for brands to focus on using visual imagery and promotions to effectively convey brand communication information. These findings highlight the importance of brand communication by emphasizing the connection between ethical brand images and consumer perceptions.

• Geophysics and Geophysical Exploration
• /
• v.1 no.1
• /
• pp.49-56
• /
• 1998

Seismic tomography has been widely used as high resolution subsurface imaging techniques in engineering applications. Although most of the techniques have been using travel time inversion, waveform method is being driven forward owing to the progress of computational environments. Although full-waveform inversion method has been known as the best method in terms of model resolving power without high-frequency restriction and weak scattering approximation, it has practical disadvantage that it is apt to get stuck in local minimum if the initial guess is far from the actual model and it consumes so much time to calculate. In this study, 2-D full-waveform inversion algorithm in acoustic medium is developed, which uses result of traveltime tomography as initial model. From the application on synthetic data, it is proved that this approach can efficiently reduce the problem of conventional approaches: our algorithm shows much faster convergence rate and improvement of model resolution. Result of application on physical modeling data also shows much improvement. It is expected that this algorithm can be applicable to real data.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.4
• /
• pp.505-516
• /
• 2018

In order to establish an efficient bridge maintenance strategy, the future performance of a bridge must be estimated by considering the current performance, which allows more rational way of decision-making in the prediction model with higher accuracy. However, personnel-based existing maintenance may result in enormous maintenance costs since it is difficult for a bridge administrator to estimate the bridge performance exactly at a targeting management level, thereby disrupting a rational decision making for bridge maintenance. Therefore, in this work, we developed a representative performance prediction model for each bridge element considering uncertainty using domestic bridge inspection data, and proposed a bayesian updating method that can apply the developed model to actual maintenance bridge with higher accuracy. Also, the feasibility analysis based on calculation of maintenance cost for monitoring maintenance scenario case is performed to propose advantages of the Bayesian-updating-driven preventive maintenance in terms of the cost efficiency in contrast to the conventional periodic maintenance.

• Proceedings of KOSOMES biannual meeting
• /
• 2003.05a
• /
• pp.191-200
• /
• 2003

Gunsan coastal area is one of region increasing pollution problems. One of the most important factors that cause eutrophication is nutrient materials containing nitrogen and phosphorus which stem from excreation of terrestial sources. At this study, the three-dimensional numerical hydrodynamic and ecosystem model, which was developed by Institute for Resources and Environment of Japan, were applied to analyze the processes affecting the eutrophication. The residual currents, which were obtained by integrating the simulated tidal currents over 1 tidal cycle, showed the presence of a typical. Density driven currents were generated westward at surface and eastward at the bottom in Geum estuary area where the fresh waters are flowing into. The ecosystem model was calibrated with the data surveyed in the field of the study area in annual average. The simulated results of DIN were fairly good coincided with the observed values within relative error of 32.39%. correlation coefficient(r) of 0.99. In the case of DIP, the simulated results were fairly good coincided with the observed values within relative error of 24.26%, correlation coefficient (r) of 0.82. The simulations of DIN and DIP concentrations were performed using ecosystem model under the conditions of 20 ∼ 80% pollution load reductions from pollution sources. In study area, concentration of DIN and DIP were reduced to 20∼80% and under 10% in case of the 80% reduction of the input loads from fresh water respectively. But pollution loads from sediment had hardly affected DIN and DIP concentration. For the environment management of coastal areas, in case of Kunsan area, the most important pollution sources affecting eutrophication phenomenon were found to be the input loads from fresh water.

• 한국정보교육학회:학술대회논문집
• /
• 2011.01a
• /
• pp.211-217
• /
• 2011

Recent IT (information technology) environmental changes, such as emerging social network services or increasing user participation in multimedia environment, have made it necessary for e-learning systems to undergo changes in various ways. Metadata is an agreement for interoperability between different systems. The standardization of metadata for e-learning system has been driven by some domestic and international organizations, but applying diverse environmental changes into the design of e-learning metadata is in dire need. In this paper, we present a methodology for the analysis and design of modeling e-learning metadata and elicit the design requirements, on the basis of the metadata standard KEM 3.0, about the elements that are expected to be needed in future e-learning systems. Based on the requirements from the analysis, we present the three-layer model for classifying the requirements by the importance of metadata elements per Kana Model. An intelligent e-learning system is to be developed based on the proposed modeling design, which we hope to influence the development of an international standard in the future.