• Korean Journal of Remote Sensing
• /
• v.38 no.5_2
• /
• pp.707-723
• /
• 2022

Although satellite-based sea surface temperature (SST) is advantageous for monitoring large areas, spatiotemporal data gaps frequently occur due to various environmental or mechanical causes. Thus, it is crucial to fill in the gaps to maximize its usability. In this study, daily SST composite fields with a resolution of 4 km were produced through a two-step machine learning approach using polar-orbiting and geostationary satellite SST data. The first step was SST reconstruction based on Data Interpolate Convolutional AutoEncoder (DINCAE) using multi-satellite-derived SST data. The second step improved the reconstructed SST targeting in situ measurements based on light gradient boosting machine (LGBM) to finally produce daily SST composite fields. The DINCAE model was validated using random masks for 50 days, whereas the LGBM model was evaluated using leave-one-year-out cross-validation (LOYOCV). The SST reconstruction accuracy was high, resulting in R² of 0.98, and a root-mean-square-error (RMSE) of 0.97℃. The accuracy increase by the second step was also high when compared to in situ measurements, resulting in an RMSE decrease of 0.21-0.29℃ and an MAE decrease of 0.17-0.24℃. The SST composite fields generated using all in situ data in this study were comparable with the existing data assimilated SST composite fields. In addition, the LGBM model in the second step greatly reduced the overfitting, which was reported as a limitation in the previous study that used random forest. The spatial distribution of the corrected SST was similar to those of existing high resolution SST composite fields, revealing that spatial details of oceanic phenomena such as fronts, eddies and SST gradients were well simulated. This research demonstrated the potential to produce high resolution seamless SST composite fields using multi-satellite data and artificial intelligence.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.381-394
• /
• 2016

Graphical model represents conditional dependencies between variables as a graph with nodes and edges. It is widely used in various fields including physics, economics, and biology to describe complex association. Conditional dependencies can be estimated from a inverse covariance matrix, where zero off-diagonal elements denote conditional independence of corresponding variables. This paper proposes a efficient BCDR (block coordinate descent with random permutation) algorithm using graphics processing units and random permutation for the CONCORD (convex correlation selection method) based on the BCD (block coordinate descent) algorithm, which estimates a inverse covariance matrix based on pseudo-likelihood. We conduct numerical studies for two network structures to demonstrate the efficiency of the proposed algorithm for the CONCORD in terms of computation times.

• Earthquakes and Structures
• /
• v.14 no.4
• /
• pp.313-322
• /
• 2018

The present study is dedicated to investigate the SH body-as well as Love-waves propagation effects in porous media with uncertain porosity and permeability. A unified formulation of the governing equations for one-dimensional (1-D) wave propagation in anisotropic porous layered media is presented deterministically. The uncertainties around the above two cited parameters are taken into account by random fields with the help of Monte Carlo Simulations (MCS). Random samples of the porosity and the permeability are generated according to the normal and lognormal distribution functions, respectively, with a mean value and a coefficient of variation for each one of the two parameters. After performing several thousands of samples, the mathematical expectation (mean) of the solution of the wave propagation equations in terms of amplification functions for SH waves and in terms of dispersion equation for Love-waves are obtained. The limits of the Love wave velocity in a porous soil layer overlaying a homogeneous half-space are obtained where it is found that random variations of porosity change the zeros of the wave equation. Also, the increase of uncertainties in the porosity (high coefficient of variation) decreases the mean amplification function amplitudes and shifts the fundamental frequencies. However, no effects are observed on both Love wave dispersion and amplification function for random variations of permeability. Lastly, the present approach is applied to a case study in the Adapazari town basin so that to estimate ground motion accelerations lacked in the fast-growing during the main shock of the damaging 1999 Kocaeli earthquake.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.60 no.3
• /
• pp.63-70
• /
• 2018

In this study, a simple probabilistic approach using equivalent coefficient of consolidation ($c_e$) was proposed to consider the spatial variability of coefficient of vertical consolidation ($c_v$), and the effect of the probability distribution of coefficient of consolidation on degree of consolidation in heterogeneous soil was investigated. The statistical characteristics of consolidation coefficient were estimated from 1,226 field data, and four probability distributions (Normal, Log-normal, Gamma, and Weibull) were applied to consider the effect of probability distribution. The random fields of coefficient of consolidation were generated based on Karhunen-Loeve expansion. Then, the equivalent coefficient of consolidation was calculated from the random field and used as the input value of consolidation analysis. As a result, the probabilistic analysis can be performed effectively by separating random field and numerical analysis, and probabilistic analysis was performed using a Latin hypercube Monte Carlo simulation. The results showed that the statistical properties of $c_e$ were changed by the probability distribution and spatial variability of $c_v$, and the probability distribution of $c_v$ has considerable effects on the probabilistic results. There was a large difference of failure probability depend on the probability distribution when the autocorrelation distance was small (i.e., highly heterogeneous soil). Therefore, the selection of a suitable probability distribution of $c_v$ is very important for reliable probabilistic analysis of consolidation.

• The Korean Journal of Applied Statistics
• /
• v.31 no.2
• /
• pp.287-301
• /
• 2018

It is common to encounter count data with excess zeros in various research fields such as the social sciences, natural sciences, medical science or engineering. Such count data have been explained mainly by zero-inflated Poisson model and extended models. Zero-inflated count data are also often correlated or clustered, in which random effects should be taken into account in the model. Frequentist approaches have been commonly used to fit such data. However, a Bayesian approach has advantages of prior information, avoidance of asymptotic approximations and practical estimation of the functions of parameters. We consider a Bayesian zero-inflated Poisson regression model with random effects for correlated zero-inflated count data. We conducted simulation studies to check the performance of the proposed model. We also applied the proposed model to smoking behavior data from the Regional Health Survey (2015) of the Korea Centers for disease control and prevention.

• Journal of the Korean Geotechnical Society
• /
• v.18 no.2
• /
• pp.97-105
• /
• 2002

Random properties of discontinuities were attributed to the limitation of test methods and lack of obtained data. Therefore, the uncertainties are pervasive and inevitable in rock slope engineering as well as other geotechnical engineering fields. The probabilistic analysis has been proposed to deal properly with the uncertainty. However, previous probabilistic approaches do not take account of the condition of kinematic instability but consider only kinetic instability. In this study, in order to overcome the limitation of the previous studies, the geometric characteristics as well as the shear strength characteristics in discontinuities are taken account into the probabilistic analysis. Then, the new approach to evaluate the probability of failure is suggested. The results of the deterministic analysis which was carried out to compare with the result of the probabilistic analysis, are somewhat different from those of the probabilistic approach. This is because the selected and used data in the deterministic approach do not take account of the random properties of discontinuities.

• Economic and Environmental Geology
• /
• v.33 no.4
• /
• pp.295-307
• /
• 2000

Probabilistic analysis is a powerful method to quantify variability and uncertainty common in engineering geology fields. In rock slope engineering, the uncertainty and variation may be in the form of scatter in orientations and geometries of discontinuities, and also test results. However, in the deterministic analysis, the factor of safety which is used to ensure stability of rock slopes, is based on the fixed representative values for each parameter without a consideration of the scattering in data. For comparison, in the probabilistic analysis, these discontinuity parameters are considered as random variables, and therefore, the reliability and probability theories are utilized to evaluate the possibility of slope failure. Therefore, in the probabilistic analysis, the factor of safety is considered as a random variable and replaced by the probability of failure to measure the level of slope stability. In this study, the stochastic properties of discontinuity parameters are evaluated and the stability of rock slope is analyzed based on the random properties of discontinuity parameters. Then, the results between the deterministic analysis and the probabilistic analysis are compared and the differences between the two analysis methods are explained.

• Annual Conference on Human and Language Technology
• /
• 2006.10e
• /
• pp.268-272
• /
• 2006

질의응답 시스템은 사용자 질의에 해당하는 정답을 찾기 위해서 세부 분류된 개체명을 사용한다. 이러한 세부 분류 개체명 인식을 위해서 대부분의 시스템이 일반 대분류 개체명인식 후에 사전 등을 이용하여 세부 분류로 나누는 방법을 이용하고 있다. 본 논문에서는 질의응답 시스템을 위한 세부 분류 개체명 인식을 위해서 Conditional Random Fields를 이용한다. 개체명 인식의 과정을 개체명 경계 인식과 경계가 인식된 개체명의 클래스 분류의 두 단계로 나누어, 개체명 경계 인식에 Conditional Random Fields를 이용하고, 경계 인식된 개체명의 클래스 분류에는 Maximum Entropy를 이용한다. 실험결과 147개의 세부분류 개체명 인식에 대해서 정확도 85.8%, 재현률 81.1%. F1=83.4의 성능을 얻었고. baseline model 보다 학습 시간이 27%로 줄고 성능은 증가하였다. 또한 제안된 세부 분류개체명 인식기를 이용하여 질의응답 시스템에 적용한 결과 26%의 성능향상을 보였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.722-728
• /
• 2020

Digital image processing has been applied in a wide range of fields due to the development of IoT technology and plays an important role in data processing. Various techniques have been proposed to remove such noise, but the conventional impulse noise canceling methods are insufficient to remove noise of edge components of an image, and have a disadvantage of being greatly affected by random impulse noise. Therefore, in this paper, we propose an algorithm that effectively removes edge component noise in random impulse noise environment. The proposed algorithm calculates the threshold value by determining the noise level and switches the filtering process by comparing the reference value with the input pixel value. The proposed algorithm shows good performance in the existing method, and the simulation results show that the noise is effectively removed from the edge of the image.

• Journal of the Korean Geotechnical Society
• /
• v.27 no.10
• /
• pp.93-104
• /
• 2011

In this paper, a numerical procedure of probabilistic steady seepage analysis that considers the spatial variability of soil permeability is presented. The procedure extends the deterministic analysis based on the finite element method to a probabilistic approach that accounts for the uncertainties and spatial variation of the soil permeability. Two-dimensional random fields are generated based on a Karhunen-Lo$\grave{e}$ve expansion in a fashion consistent with a specified marginal distribution function and an autocorrelation function. A Monte Carlo simulation is then used to determine the statistical response based on the random fields. A series of analyses were performed to verify the application potential of the proposed method and to study the effects of uncertainty due to the spatial heterogeneity on the seepage behavior of soil foundation beneath water retaining structure with a single sheet pile wall. The results showed that the probabilistic framework can be used to efficiently consider the various flow patterns caused by the spatial variability of the soil permeability in seepage assessment for a soil foundation beneath water retaining structures.