• Health Policy and Management
• /
• v.28 no.4
• /
• pp.402-410
• /
• 2018

Background: Monitoring appropriate medication categories can provide early warning of certain disease outbreaks. This study aimed to present a methodology for selecting and monitoring medications relevant to the surveillance of acute respiratory tract infections, such as influenza. Methods: To estimate correlations between acute febrile respiratory tract infection and some medication categories, the cross-correlation coefficient (CCC) was used and established. Two databases were used: real-time prescription trend of antivirals, anti-inflammatory drugs, antibiotics using Drug Utilization Review Program between 2012 and 2015 and physicians' number of encounters with acute febrile respiratory tract infections such as influenza outbreaks using the national level health insurance claims data. The seasonality was also evaluated using the CCC. Results: After selecting six candidate diseases that require extensive monitoring, influenza with highly specific medical treatment according to the health insurance claims data and its medications were chosen as final candidates based on a data-driven approach. Antiviral medications and influenza were significantly correlated. Conclusion: An annual correlation was observed between influenza and antiviral medications, anti-inflammatory drugs. Suitable models should be established for syndromic surveillance of influenza.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.4
• /
• pp.7-16
• /
• 2023

For decades, creating a desired locomotive motion in a goal-oriented manner has been a challenge in character animation. Data-driven methods using generative models have demonstrated efficient ways of predicting long sequences of motions without the need for explicit conditioning. While these methods produce high-quality long-term motions, they can be limited when it comes to synthesizing motion for challenging novel scenarios, such as punching a random target. A state-of-the-art solution to overcome this limitation is by using a GAN Discriminator to imitate motion data clips and incorporating reinforcement learning to compose goal-oriented motions. In this paper, our research aims to create characters performing combat sports such as boxing, using a novel reward design in conjunction with existing GAN-based approaches. We experimentally demonstrate that both the Adversarial Motion Prior [3] and Adversarial Skill Embeddings [4] methods are capable of generating viable motions for a character punching a random target, even in the absence of mocap data that specifically captures the transition between punching and locomotion. Also, with a single learned policy, multiple task controllers can be constructed through the TimeChamber framework.

• Journal of Korean Society of Water and Wastewater
• /
• v.37 no.6
• /
• pp.347-361
• /
• 2023

Water utilities are making various efforts to reduce water losses from water networks, and an essential part of them is to recognize the moment when a pipe burst occurs during operation quickly. Several physics-based methods and data-driven analysis are applied using real-time flow and pressure data measured through a SCADA system or smart meters, and methodologies based on machining learning are currently widely studied. Water utilities should apply various approaches together to increase pipe burst detection. The most intuitive and explainable water balance method and its procedure were presented in this study, and the applicability and detection performance were evaluated by applying this approach to water supply pipelines. Based on these results, water utilities can establish a mass balance-based pipe burst detection system, give a guideline for installing new flow meters, and set the detection parameters with expected performance. The performance of the water balance analysis method is affected by the water network operation conditions, the characteristics of the installed flow meter, and event data, so there is a limit to the general use of the results in all sites. Therefore, water utilities should accumulate experience by applying the water balance method in more fields.

• Spatial Information Research
• /
• v.1 no.1
• /
• pp.39-53
• /
• 1993

A geographical information systems(GIS) was a useful aid in the assessment of urban nonpoint source pollution and the development of a pollution control strategy. The GIS was used for data integration and display, and to provide data for a nonpoint source model. An empirical nonpoint source loading model driven by land use was used to estimate pollutant loadings of priority pollutant. Pollutant loadings were estimated at fine spatial resolution and aggregated to storm sewer drainage basins(sewershedsl. Eleven sewersheds were generated from digital versions of sewer maps. The pollutant loadings of individual land use polygons, derived as the unit of analysis from street blocks, were aggregated to get total pollutant loading within each sewershed. Based on the model output, a critical sewershed was located. Pollutant loadings at major sewer junctions within the critical sewershed were estimated to develop a mi t igat ion strategy. Two approaches based on the installat ion of wet ponds were investigated -- a regional approach using one large wet pond at the major sewer outfall and a multi-site approach using a number of smaller sites for each major sewer junction. Cost analysis showed that the regional approach would be more cost effective, though it would provide less pollution control.

• Journal of the Korean Geographical Society
• /
• v.43 no.1
• /
• pp.134-153
• /
• 2008

The main purpose of the research is to illustrate the value of the spatial statistical approach to residential differentiation by providing a framework for exploratory spatial data analysis (ESDA) using a local spatial separation measure. ESDA aims, by utilizing a variety of statistical and cartographic visualization techniques, at seeking to detect patterns, to formulate hypotheses, and to assess statistical models for spatial data. The research is driven by a realization that ESDA based on local statistics has a great potential for substantive research. The main results are as follows. First, a local spatial separation measure is correspondingly derived from its global counterpart. Second, a set of significance testing methods based on both total and conditional randomization assumptions is provided for the local measure. Third, two mapping techniques, a 'spatial separation scatterplot map' and a 'spatial separation anomaly map', are devised for ESDA utilizing the local measure and the related significance tests. Fourth, a case study of residential differentiation between the highly educated and the least educated in major Korean metropolitan cities shows that the proposed ESDA techniques are beneficial in identifying bivariate spatial clusters and spatial outliers.

• Korean Chemical Engineering Research
• /
• v.62 no.3
• /
• pp.214-224
• /
• 2024

Valves play an essential role in a chemical plant such as regulating fluid flow and pressure. Therefore, optimal selection of the valve size and type is essential task. Valve size and type have been selected based on theoretical formulas about calculating valve sizing coefficient (Cv). However, this approach has limitations such as requiring expert knowledge and consuming substantial time and costs. Herein, this study developed a model for predicting valve sizes and types using machine learning. We developed models using four algorithms: ANN, Random Forest, XGBoost, and Catboost and model performances were evaluated using NRMSE & R² score for size prediction and F1 score for type prediction. Additionally, a case study was conducted to explore the impact of phases on valve selection, using four datasets: total fluids, liquids, gases, and steam. As a result of the study, for valve size prediction, total fluid, liquid, and gas dataset demonstrated the best performance with Catboost (Based on R², total: 0.99216, liquid: 0.98602, gas: 0.99300. Based on NRMSE, total: 0.04072, liquid: 0.04886, gas: 0.03619) and steam dataset showed the best performance with RandomForest (R²: 0.99028, NRMSE: 0.03493). For valve type prediction, Catboost outperformed all datasets with the highest F1 scores (total: 0.95766, liquids: 0.96264, gases: 0.95770, steam: 1.0000). In Engineering Procurement Construction industry, the proposed fluid-specific machine learning-based model is expected to guide the selection of suitable valves based on given process conditions and facilitate faster decision-making.

• Korean Journal of Remote Sensing
• /
• v.15 no.2
• /
• pp.91-105
• /
• 1999

Spatial data integration using multiple geo-based data sets has been regarded as one of the primary GIS application issues. As for this issue, several integration schemes have been developed as the perspectives of mathematical geology or geo-mathematics. However, research-based approaches for statistical/quantitative assessments between integrated layer and input layers are not fully considered yet. Related to this niche point, in this study, spatial data integration using multiple geoscientific data sets by known integration algorithms was primarily performed. For spatial integration by using raster-based GIS functionality, geological, geochemical, geophysical data sets, DEM-driven data sets and remotely sensed imagery data sets from the Ogdong area were utilized for geological thematic mapping related by mineral potential mapping. In addition, statistical/quantitative information extraction with respective to relationships among used data sets and/or between each data set and integrated layer was carried out, with the scope of multiple data fusion and schematic statistical assessment methodology. As for the spatial integration scheme, certainty factor (CF) estimation and principal component analysis (PCA) were applied. However, this study was not aimed at direct comparison of both methodologies; whereas, for the statistical/quantitative assessment between integrated layer and input layers, some statistical methodologies based on contingency table were focused. Especially, for the bias reduction, jackknife technique was also applied in PCA-based spatial integration. Through the statistic analyses with respect to the integration information in this case study, new information for relationships of integrated layer and input layers was extracted. In addition, influence effects of input data sets with respect to integrated layer were assessed. This kind of approach provides a decision-making information in the viewpoint of GIS and is also exploratory data analysis in conjunction with GIS and geoscientific application, especially handing spatial integration or data fusion with complex variable data sets.

• The Journal of Information Systems
• /
• v.30 no.1
• /
• pp.151-177
• /
• 2021

Purpose This paper aims to improve management performance by effectively responding to production needs and reducing inventory through synchronizing production planning and procurement in the aviation industry. In this study, the differences in production planning and execution were first analyzed in terms of demand, supply, inventory, and process using the big data collected from a domestic aircraft manufacturers. This paper analyzed the problems in procurement and inventory management using legacy big data from ERP system in the company. Based on the analysis, we performed a simulation to derive an efficient procurement and inventory management plan. Through analysis and simulation of operational data, we were able to discover procurement and inventory policies to effectively respond to production needs. Design/methodology/approach This is an empirical study to analyze the cause of decrease in inventory turnover and increase in inventory cost due to dis-synchronize between production requirements and procurement. The actual operation data, a total of 21,306,611 transaction data which are 18 months data from January 2019 to June 2020, were extracted from the ERP system. All them are such as basic information on materials, material consumption and movement history, inventory/receipt/shipment status, and production orders. To perform data analysis, it went through three steps. At first, we identified the current states and problems of production process to grasp the situation of what happened, and secondly, analyzed the data to identify expected problems through cross-link analysis between transactions, and finally, defined what to do. Many analysis techniques such as correlation analysis, moving average analysis, and linear regression analysis were applied to predict the status of inventory. A simulation was performed to analyze the appropriate inventory level according to the control of fluctuations in the production planing. In the simulation, we tested four alternatives how to coordinate the synchronization between the procurement plan and the production plan. All the alternatives give us more plausible results than actual operation in the past. Findings Based on the big data extracted from the ERP system, the relationship between the level of delivery and the distribution of fluctuations was analyzed in terms of demand, supply, inventory, and process. As a result of analyzing the inventory turnover rate, the root cause of the inventory increase were identified. In addition, based on the data on delivery and receipt performance, it was possible to accurately analyze how much gap occurs between supply and demand, and to figure out how much this affects the inventory level. Moreover, we were able to obtain the more predictable and insightful results through simulation that organizational performance such as inventory cost and lead time can be improved by synchronizing the production planning and purchase procurement with supply and demand information. The results of big data analysis and simulation gave us more insights in production planning, procurement, and inventory management for smart manufacturing and performance improvement.

• Journal of Digital Convergence
• /
• v.10 no.6
• /
• pp.297-302
• /
• 2012

In this paper, we present a object detection and tracking method based on image context analysis. It is robust from the image variations such as complicated background, dynamic movement of the object. Image context analysis is carried out using the hybrid network of k-means and RBF. The proposed object detection employs context-driven adaptive Bayesian framework to relive the effect due to uneven object images. The proposed method used feature vector generator using 2D Haar wavelet transform and the Bayesian discriminant method in order to enhance the speed of learning. The system took less time to learn, and learning in a wide variety of data showed consistent results. After we developed the proposed method was applied to real-world environment. As a result, in the case of the object to detect pass outside expected area or other changes in the uncertain reaction showed that stable. The experimental results show that the proposed approach can achieve superior performance using various data sets to previously methods.

• Biotechnology and Bioprocess Engineering:BBE
• /
• v.10 no.5
• /
• pp.425-431
• /
• 2005

The systems-level analysis of microbes with myriad of heterologous data generated by omics technologies has been applied to improve our understanding of cellular function and physiology and consequently to enhance production of various bioproducts. At the heart of this revolution resides in silico genome-scale metabolic model, In order to fully exploit the power of genome-scale model, a systematic approach employing user-friendly software is required. Metabolic flux analysis of genome-scale metabolic network is becoming widely employed to quantify the flux distribution and validate model-driven hypotheses. Here we describe the development of an upgraded MetaFluxNet which allows (1) construction of metabolic models connected to metabolic databases, (2) calculation of fluxes by metabolic flux analysis, (3) comparative flux analysis with flux-profile visualization, (4) the use of metabolic flux analysis markup language to enable models to be exchanged efficiently, and (5) the exporting of data from constraints-based flux analysis into various formats. MetaFluxNet also allows cellular physiology to be predicted and strategies for strain improvement to be developed from genome-based information on flux distributions. This integrated software environment promises to enhance our understanding on metabolic network at a whole organism level and to establish novel strategies for improving the properties of organisms for various biotechnological applications.