• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.60-60
• /
• 2020

기후변화 및 지구온난화로 인한 자연재해 규모가 점차 대형화, 다양화되고 있어 이로 인한 피해도 증대되고 있다. 특히, 다양한 시설과 인구밀도가 높은 도심 지역은 집중호우, 태풍, 홍수 등 자연재해에 취약하여 인적·물적 피해 위험성이 매우 높다. 방재 시설확보 및 개선을 통한 더 높은 안정성 및 기상예보를 통한 대응, 대책을 통한 피해 저감이 이루어지고 있다. 그러나 일반적으로 제공되는 단일 수치모형 기반의 결정론적 기상예측정보는 기상 상태, 선행시간, 모형 매개변수 등으로 인한 불확실성이 매우 크며 이에 대한 정보가 제공되지 않다. 이러한 문제점을 보완하기 위해 앙상블 수치모델 정보와 기상레이더 자료 기반의 단기 예측정보가 활용이 가능하다. 그러나, 앙상블 수치모델의 불확실성, 기상레이더 기반 예측정보의 짧은 예측 선행시간으로 인해 수문학적 모형에 입력자료로 활용은 어려운 실점이다. 본 연구에서는 지점 관측자료의 시간적 연속성, 기상레이더 자료의 공간적 연속성, 앙상블 예측정보의 선행시간 정보를 융합하여 기상예측정보에 대한 불확실성 개선 및 선행시간에 따른 정확도를 높일 방법을 제안하였다. 기상청에서 제공하는 앙상블 예측자료인 LENS 자료, 레이더 강수량, ASOS 관측자료 기반으로 분석이 수행되었으며 분석결과는 예측강수량을 활용하는 분야에 긍정적 영향을 미칠 것으로 기대된다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.181-181
• /
• 2021

유역 내의 물순환 평가를 위하여 적합한 강우-유출모형을 선정하고 적용하는 것은 수문학적 관점에서 주된 과제이다. 장기적인 관점의 수자원 관리를 위해서는 직접적인 계측을 통해 장기간의 유출자료를 취득하는 방법이 있으나, 국내의 주요지점을 제외한 대다수의 중소규모의 지점에 계측기를 설치하는 것은 현실적으로 어려우므로, 자료취득이 비교적 용이하고 신뢰성이 높은 장기간 강우 자료를 강우-유출모형의 입력자료로 활용하여 미계측 유역으로의 모형을 확장하는 방안이 적절하다는 평가를 받고 있다. 본 연구는 국내외 주요 연속강우-유출모형의 특성을 파악하기 위하여 비교적 신뢰성 있는 자료를 보유하고 있는 소양강댐 유역에 다수의 연속강우-유출모형을 적용하였다. 모델링 결과로 산출된 유황곡선(flow duration curve)을 소양강댐 유입량과 비교하여 각 모형의 특징을 파악하고 유량에 따른 적합성 평가를 진행하였다. 또한, 향후 미계측유역으로 모형을 확장하기 위하여 매개변수 개수 및 재현능력을 동시에 평가하였다. 다수의 모형 중 적합성이 높은 모형들을 선별하였으며, 선별된 모형들의 불확실성을 고려함과 동시에 계층적 베이지안 기법을 활용하여 최종적으로 앙상블모형을 제시하였다. 앙상블모형을 단일 모형과 비교한 결과 단일 모형보다 개선된 성능을 확인하였다.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.728-735
• /
• 2022

Roofers get exposed to increased risk of knee musculoskeletal disorders (MSDs) at different phases of a sloped shingle installation task. As different phases are associated with different risk levels, this study explored the application of machine learning for automated classification of seven phases in a shingle installation task using knee kinematics and roof slope information. An optical motion capture system was used to collect knee kinematics data from nine subjects who mimicked shingle installation on a slope-adjustable wooden platform. Four features were used in building a phase classification model. They were three knee joint rotation angles (i.e., flexion, abduction-adduction, and internal-external rotation) of the subjects, and the roof slope at which they operated. Three ensemble machine learning algorithms (i.e., random forests, decision trees, and k-nearest neighbors) were used for training and prediction. The simulations indicate that the k-nearest neighbor classifier provided the best performance, with an overall accuracy of 92.62%, demonstrating the considerable potential of machine learning methods in detecting shingle installation phases from workers knee joint rotation and roof slope information. This knowledge, with further investigation, may facilitate knee MSD risk identification among roofers and intervention development.

• Asia pacific journal of information systems
• /
• v.29 no.4
• /
• pp.789-816
• /
• 2019

Text classification is a challenging task, especially when dealing with a huge amount of text data. The performance of a classification model can be varied depending on what type of words contained in the document corpus and what type of features generated for classification. Aside from proposing a new modified version of the existing algorithm or creating a new algorithm, we attempt to modify the use of data. The classifier performance is usually affected by the quality of learning data as the classifier is built based on these training data. We assume that the data from different domains might have different characteristics of noise, which can be utilized in the process of learning the classifier. Therefore, we attempt to enhance the robustness of the classifier by injecting the heterogeneous data artificially into the learning process in order to improve the classification accuracy. Semi-supervised approach was applied for utilizing the heterogeneous data in the process of learning the document classifier. However, the performance of document classifier might be degraded by the unlabeled data. Therefore, we further proposed an algorithm to extract only the documents that contribute to the accuracy improvement of the classifier.

• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.771-780
• /
• 2009

Cataract is the main cause of blindness and visual impairment, especially, age-related cataract accounts for about half of the 32 million cases of blindness worldwide. As the life expectancy and the expansion of the elderly population are increasing, the cases of cataract increase as well, which causes a serious economic and social problem throughout the country. However, the incidence of cataract can be reduced dramatically through early diagnosis and prevention. In this study, we developed a prediction model of cataracts for early diagnosis using hospital data of 3,237 subjects who received the screening test first and then later visited medical center for cataract check-ups cataract between 1994 and 2005. To develop the prediction model, we used random forests and compared the predictive performance of this model with other common discriminant models such as logistic regression, discriminant model, decision tree, naive Bayes, and two popular ensemble model, bagging and arcing. The accuracy of random forests was 67.16%, sensitivity was 72.28%, and main factors included in this model were age, diabetes, WBC, platelet, triglyceride, BMI and so on. The results showed that it could predict about 70% of cataract existence by screening test without any information from direct eye examination by ophthalmologist. We expect that our model may contribute to diagnose cataract and help preventing cataract in early stages.

• Journal of Digital Convergence
• /
• v.16 no.1
• /
• pp.231-242
• /
• 2018

The efficient management of the Length of Stay(LOS) is important in hospital. It is import to reduce medical cost for patients and increase profitability for hospitals. In order to efficiently manage LOS, it is necessary to develop an artificial intelligence-based prediction model that supports hospitals in benchmarking and reduction ways of LOS. In order to develop a predictive model of LOS for acute stroke patients, acute stroke patients were extracted from 2013 and 2014 discharge injury patient data. The data for analysis was classified as 60% for training and 40% for evaluation. In the model development, we used traditional regression technique such as multiple regression analysis method, artificial intelligence technique such as interactive decision tree, neural network technique, and ensemble technique which integrate all. Model evaluation used Root ASE (Absolute error) index. They were 23.7 by multiple regression, 23.7 by interactive decision tree, 22.7 by neural network and 22.7 by esemble technique. As a result of model evaluation, neural network technique which is artificial intelligence technique was found to be superior. Through this, the utility of artificial intelligence has been proved in the development of the prediction LOS model. In the future, it is necessary to continue research on how to utilize artificial intelligence techniques more effectively in the development of LOS prediction model.

• Journal of Korea Water Resources Association
• /
• v.49 no.9
• /
• pp.761-771
• /
• 2016

The GloSea5 (Global Seasonal forecasting system version 5) is provided and operated by the KMA (Korea Meteorological Administration). GloSea5 provides Forecast (FCST) and Hindcast (HCST) data and its horizontal resolution is about 60km ($0.83^{\circ}{\times}0.56^{\circ}$) in the mid-latitudes. In order to use this data in watershed-scale water management, GloSea5 needs spatial-temporal downscaling. As such, statistical downscaling was used to correct for systematic biases of variables and to improve data reliability. HCST data is provided in ensemble format, and the highest statistical correlation ($R^2=0.60$, RMSE = 88.92, NSE = 0.57) of ensemble precipitation was reported for the Yongdam Dam watershed on the #6 grid. Additionally, the original GloSea5 (600.1 mm) showed the greatest difference (-26.5%) compared to observations (816.1 mm) during the summer flood season. However, downscaled GloSea5 was shown to have only a -3.1% error rate. Most of the underestimated results corresponded to precipitation levels during the flood season and the downscaled GloSea5 showed important results of restoration in precipitation levels. Per the analysis results of spatial autocorrelation using seasonal Moran's I, the spatial distribution was shown to be statistically significant. These results can improve the uncertainty of original GloSea5 and substantiate its spatial-temporal accuracy and validity. The spatial-temporal reproducibility assessment will play a very important role as basic data for watershed-scale water management.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.36-42
• /
• 2021

With recent developments in the Fourth Industrial Revolution, the manufacturing industry has changed rapidly. Through key aspects of Fourth Industrial Revolution super-connections and super-intelligence, machine learning will be able to make fault predictions during the foam-making process. Polyol and isocyanate are components in polyurethane foam. There has been a lot of research that could affect the characteristics of the products, depending on the specific mixture ratio and temperature. Based on these characteristics, this study collects data from each factor during the foam-making process and applies them to machine learning in order to predict faults. The algorithms used in machine learning are the decision tree, kNN, and an ensemble algorithm, and these algorithms learn from 5,147 cases. Based on 1,000 pieces of data for validation, the learning results show up to 98.5% accuracy using the ensemble algorithm. Therefore, the results confirm the faults of currently produced parts by collecting real-time data from each factor during the foam-making process. Furthermore, control of each of the factors may improve the fault rate.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1427-1435
• /
• 2023

Vegetation indices based on selected wavelength reflectance measurements are used to represent crop growth and physiological conditions. However, the anisotropic properties of the crop canopy surface can govern spectral reflectance and vegetation indices. In this study, we applied an ensemble of bidirectional reflectance distribution function (BRDF) models to high-resolution Sentinel-2 satellite imagery and compared the differences between correction results before and after reflectance. In the red and near-infrared (NIR) band reflectance images, BRDF-corrected outlier values appeared in certain urban and paddy fields of farmland areas and forest shadow areas. These effects were equally observed when calculating the normalized difference vegetation index (NDVI) and 2-band enhanced vegetation index (EVI2). Furthermore, the outlier values in corrected NIR band were shown in pixels shadowed by mountain terrain. These results are expected to contribute to the development and improvement of BRDF models in high-resolution satellite images.

• Journal of Korea Water Resources Association
• /
• v.57 no.9
• /
• pp.571-583
• /
• 2024

Forecasting dam inflows in the medium to long term is crucial for effective dam operation and the prevention of water-related disasters such as floods and droughts. However, the increasing frequency of extreme weather events due to climate change has made hydrological forecasting more challenging. Since 2000, seasonal weather forecasts, which provide predictions for weather variables up to about seven months ahead, and their hydrological interpretation, known as Seasonal Flow Forecasts (SFFs) have gained significant global interest. This study utilises seasonal weather forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF), converting them into inflow forecasts using a hydrological model for 12 multipurpose dams in South Korea from 2011 to 2020. We then compare the performance of these SFFs with the Ensemble Streamflow Prediction (ESP). Our results indicate that while SFFs are more effective for short-term predictions of 1-2 months, ESP outperforms SFFs for long-term predictions. Seasonally, the performance of SFFs is higher in October-November but lower from December to February. Moreover, our findings demonstrate that SFFs are highly effective in quantitatively predicting dry conditions, although they tend to underestimate inflows under wet conditions.