• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.170-170
• /
• 2017

본 연구의 목표는 극한 지역의 대비 시스템을 구축하기 위하여 인공 신경망(Artificial Neural Networks)을 이용하여 보다 관측하기 쉬운 기상 인자들로부터 적설량을 실시간 측정 가능성을 제시하는 것이다. 본 연구에서 사용한 데이터베이스는 기상청의 기상자료개방포털에서 사람이 직접 측정한 종관기상관측의 자료다. 이 중에서 일최대 기온, 일최저 기온, 일평균 기온, 강수량을 사용하여 오차를 줄여나가는 최적화방법으로 인공 신경망 시스템을 설계하였다. 설계된 시스템으로 500회 시뮬레이션한 연구 결과는 상관계수가 적설량 측정에 대한 인공 신경망의 크기(노드의 개수)와 관계없이 평균적으로 0.8627인 것을 보여준다. 추가적으로 보조 입력 값인 고도를 사용한 결과, 성능은 좋아졌지만 상관계수의 차이는 평균 0.0044로 미세했다. 또한 Cross-Validation을 통해 기존의 보간법인 Kriging기법과 비교하여 미 관측 지역에서 인공 신경망(ANNs) 사용이 Kriging기법 보다 우수하다는 것을 2차원 Regression's map을 통해 나타냈다. 마지막으로 오차가 크게 발생했을 경우 보안할 수 있는 확률적인 방안을 제시하였다.

• Journal of the Korean Society for Heat Treatment
• /
• v.34 no.5
• /
• pp.211-217
• /
• 2021

To predict mechanical properties of secondary hardening martensitic steels, a machine learning ensemble model was established. Based on ANN(Artificial Neural Network) architecture, some kinds of methods was considered to optimize the model. In particular, interaction features, which can reflect interactions between chemical compositions and processing conditions of real alloy system, was considered by means of feature engineering, and then K-Fold cross validation coupled with bagging ensemble were investigated to reduce R2_score and a factor indicating average learning errors owing to biased experimental database.

• International Journal of Contents
• /
• v.17 no.4
• /
• pp.79-90
• /
• 2021

With the deepening of population aging, pension has become an urgent problem in most countries. Community smart pension can effectively resolve the problem of traditional pension, as well as meet the personalized and multi-level needs of the elderly. To predict the pension intention of the elderly in the community more accurately, this paper uses the decision tree classification method to classify the pension data. After missing value processing, normalization, discretization and data specification, the discretized sample data set is obtained. Then, by comparing the information gain and information gain rate of sample data features, the feature ranking is determined, and the C4.5 decision tree model is established. The model performs well in accuracy, precision, recall, AUC and other indicators under the condition of 10-fold cross-validation, and the precision was 89.5%, which can provide the certain basis for government decision-making.

• International Journal of Advanced Culture Technology
• /
• v.10 no.1
• /
• pp.236-241
• /
• 2022

We consider methods of estimating a binary regression function using a nonparametric kernel estimation when there is only one covariate. For this, the Nadaraya-Watson estimation method using single and double bandwidths are used. For choosing a proper smoothing amount, the cross-validation and plug-in methods are compared. In the real data analysis for case study, German credit data and heart disease data are used. We examine whether the nonparametric estimation for binary regression function is successful with the smoothing parameter using the above two approaches, and the performance is compared.

• Proceedings of the Korean Society for Applied Microbiology Conference
• /
• 2000.04a
• /
• pp.186-192
• /
• 2000

• Nuclear Engineering and Technology
• /
• v.53 no.10
• /
• pp.3151-3157
• /
• 2021

Copper is an important structural material in various nuclear energy applications, therefore the correct knowledge of copper cross sections is crucial. The presented paper deals with a validation of different copper transport libraries by means of activation of selected samples. An intense ²⁵²Cf(sf) source with a reference neutron spectrum was used as a neutron source. After irradiation, the samples were measured using a high purity germanium detector and the dosimeter reaction rates were inferred. These experimental data were compared with MCNP6 calculations using CENDL-3.1, JENDL-4.0, ENDF/B-VII.1, ENDF/B-VIII.0, JEFF-3.2 and JEFF-3.3 evaluated Cu transport libraries. The experiment specifically focuses on ⁵⁸Ni(n,p)⁵⁸Co, ⁹³Nb(n,2n)^92mNb, ¹⁹⁷Au(n,g)¹⁹⁸Au and ⁵⁵Mn(n,g)⁵⁶Mn dosimetry reactions. Evaluated activation cross sections of these dosimetric reactions were taken from the IRDFF-II library. The best library performance depends on the energy region of interest.

• Nuclear Engineering and Technology
• /
• v.52 no.12
• /
• pp.2717-2724
• /
• 2020

Silicon, especially silicon in the form of SiO₂, is a major component of rocks. Final spent fuel storages, which are being designed, are located in suitable rock formations in the Earth's crust. Reduction of the uncertainty of silicon neutron scattering and capture is needed; improved silicon evaluations have been recently produced by the ORNL/IAEA collaboration within the INDEN project. This paper deals with the nuclear data validation of that evaluation performed at the LR-0 reactor by means of critical experiments and measurement of reaction rates. Large amounts of silicon were used both as pure crystalline silicon and SiO₂ sand. The critical moderator level was measured for various core configurations. Reaction rates were determined in the largest core configuration. Simulations of the experimental setup were performed using the MCNP6.2 code. The obtained results show the improvement in silicon cross-sections in the INDEN evaluations compared to existing evaluations in major libraries. The new Thermal Scattering Law for SiO₂ published in ENDF/B-VIII.0 additionally reduces the discrepancy between calculation and experiments. However, an unphysical peak is visible in the neutron spectrum in SiO₂ obtained by calculation with the new Thermal Scattering Law.

• Journal of Korea Water Resources Association
• /
• v.52 no.12
• /
• pp.963-973
• /
• 2019

This study reviews a urban flooding risk criteria estimation model to predict risk criteria in areas where flood risk criteria are not precalculated by using watershed characteristic data and limit rainfall based on damage history. The risk criteria estimation model was designed using Support Vector Machine, one of the machine learning algorithms. The learning data consisted of regional limit rainfall and watershed characteristic. The learning data were applied to the SVM algorithm after normalization. We calculated the mean absolute error and standard deviation using Leave-One-Out and K-fold cross-validation algorithms and evaluated the performance of the model. In Leave-One-Out, models with small standard deviation were selected as the optimal model, and models with less folds were selected in the K-fold. The average accuracy of the selected models by rainfall duration is over 80%, suggesting that SVM can be used to estimate flooding risk criteria.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.12
• /
• pp.2511-2520
• /
• 2011

This paper describes the parameter adjustment method of fuzzy membership function to improve the performance of multi-sensor fusion system using adaptive fuzzy-Kalman filter and cross-validation. The adaptive fuzzy-Kanlman filter has two input parameters, variation of accelerometer measurements and residual error of Kalman filter. The filter estimates system noise R and measurement noise Q, then changes the Kalman gain. To evaluate proposed adaptive fuzzy-Kalman filter, we make the two-axis AHRS(Attitude Heading Reference System) using fusion of an accelerometer and a gyro sensor. Then we verified its performance by comparing to NAV420CA-100 to be used in various fields of airborne, marine and land applications.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.6
• /
• pp.1113-1125
• /
• 2013

Corporate bankruptcy prediction has been an important topic in the accounting and finance field for a long time. Several data mining techniques have been used for bankruptcy prediction. However, there are many limits for application to real classification problem with a single model. This study proposes ensemble SVM (support vector machine) model which assembles different SVM models with each different kernel functions. Our ensemble model is made and evaluated by v-fold cross-validation approach. The k top performing models are recruited into the ensemble. The classification is then carried out using the majority voting opinion of the ensemble. In this paper, we investigate the performance of ensemble SVM classifier in terms of accuracy, error rate, sensitivity, specificity, ROC curve, and AUC to compare with single SVM classifiers based on financial ratios dataset and simulation dataset. The results confirmed the advantages of our method: It is robust while providing good performance.