• Structural Monitoring and Maintenance
• /
• v.9 no.2
• /
• pp.179-200
• /
• 2022

This paper proposes a data-driven methodology for online early damage identification under changing environmental conditions. The proposed method relies on two data analysis methods: feature-based method and hybrid principal component analysis (PCA) and kernel PCA to separate damage from environmental influences. First, spectral sub-band features, namely, spectral sub-band centroids (SSCs) and log spectral sub-band energies (LSSEs), are proposed as damage-sensitive features to extract damage information from measured structural responses. Second, hybrid modeling by integrating PCA and kernel PCA is performed on the spectral sub-band feature matrix for data normalization to extract both linear and nonlinear features for nonlinear procedure monitoring. After feature normalization, suppressing environmental effects, the control charts (Hotelling T² and SPE statistics) is implemented to novelty detection and distinguish damage in structures. The hybrid PCA-KPCA technique is compared to KPCA by applying support vector machine (SVM) to evaluate the effectiveness of its performance in detecting damage. The proposed method is verified through numerical and full-scale studies (a Bridge Health Monitoring (BHM) Benchmark Problem and a cable-stayed bridge in China). The results demonstrate that the proposed method can detect the structural damage accurately and reduce false alarms by suppressing the effects and interference of environmental variations.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.23-26
• /
• 2023

본 논문에서는 서술형 수학 문제 풀이 모델의 숫자 대소관계 파악을 위한 명시적 자질추출방식 Explicit Feature Extraction(EFE) Reasoner 모델을 제안한다. 서술형 수학 문제는 자연현상이나 일상에서 벌어지는 사건을 수학적으로 기술한 문제이다. 서술형 수학 문제 풀이를 위해서는 인공지능 모델이 문장에 함축된 논리를 파악하여 수식 또는 답을 도출해야 한다. 때문에 서술형 수학 문제 데이터셋은 인공지능 모델의 언어 이해 및 추론 능력을 평가하는 지표로 활용되고 있다. 기존 연구에서는 문제를 이해할 때 숫자의 대소관계를 파악하지 않고 문제에 등장하는 변수의 논리적인 관계만을 사용하여 수식을 도출한다는 한계점이 존재했다. 본 논문에서는 자연어 이해계열 모델 중 SVAMP 데이터셋에서 가장 높은 성능을 내고 있는 Deductive-Reasoner 모델에 숫자의 대소관계를 파악할 수 있는 방법론인 EFE 를 적용했을 때 RoBERTa-base 에서 1.1%, RoBERTa-large 에서 2.8%의 성능 향상을 얻었다. 이 결과를 통해 자연어 이해 모델이 숫자의 대소관계를 이해하는 것이 정답률 향상에 기여할 수 있음을 확인한다.

• Communications for Statistical Applications and Methods
• /
• v.19 no.4
• /
• pp.619-628
• /
• 2012

Noise reduction is an important issue in the field of image processing because image noise lowers the quality of the original pure image. The basic difficulty is that the noise and the signal are not easily distinguished. Simple smoothing is the most basic and important procedure to effectively remove the noise; however, the weakness is that the feature area is simultaneously blurred. In this research, we use ways to measure the degree of noise with respect to the degree of image features and propose a Bayesian noise reduction method based on MAP (maximum a posteriori). Simulation results show that the proposed adaptive noise reduction algorithm using Bayesian MAP provides good performance regardless of the level of noise variance.

• Communications for Statistical Applications and Methods
• /
• v.12 no.3
• /
• pp.557-587
• /
• 2005

This paper elucidates the limiting Gaussian distribution of a class of rank order statistics {$T_N$} for two sample problem pertaining to empirical processes of the squared residuals from two independent samples of GARCH processes. A distinctive feature is that, unlike the residuals of ARMA processes, the asymptotics of {$T_N$} depend on those of GARCH volatility estimators. Based on the asymptotics of {$T_N$}, we empirically assess the relative asymptotic efficiency and effect of the GARCH specification for some GARCH residual distributions. In contrast with the independent, identically distributed or ARMA settings, these studies illuminate some interesting features of GARCH residuals.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.745-758
• /
• 2011

An Economic Sentiment Indicator(ESI) is a composite indicator of business survey indices(BSI) and consumer survey indices(CSI). The ESI designed to reflect economic agents' (this includes producers and consumers) overall perceptions of economic activity in a one-dimensional index. The European Commission has published an ESI since 1985. This paper demonstrates the construction of an ESI for the Korean economy. The BSI and CSI components (having a high correlation and a leading feature with respect to GDP) are selected to construct the ESI and they are aggregated using a weighted average and then scaled to have a long-term average of 100 and a standard deviation of 10. Thus values greater than 100 indicate an above-average economic sentiment and vice versa. The newly constructed Korean ESI that extends to January 2003 shows a good tracking performance of GDP and adequately reflects the overall perception of economic activity.

• The Korean Journal of Applied Statistics
• /
• v.24 no.6
• /
• pp.1249-1257
• /
• 2011

In this paper, we propose a Sequential Adaptive Nearest Neighbor(SANN) imputation method that combines the Adaptive Nearest Neighbor(ANN) method and the Sequential k-Nearest Neighbor(SKNN) method. When choosing the nearest neighbors of missing observations, the proposed SANN method takes the local feature of the missing observations into account as well as reutilizes the imputed observations in a sequential manner. By using a Monte Carlo study and a real data example, we demonstrate the characteristics of the SANN method and its potential performance.

• Journal of the Korean Statistical Society
• /
• v.22 no.2
• /
• pp.159-169
• /
• 1993

In this article, we consider two sample problem with randomly right censored data. We propse two-sample confidence intervals for the difference in medians or any quantiles, based on bootstrap. The bootstrap version of two-sample confidence intervals proposed in this article is simple to apply and do not need the assumption of the shift model, so that for the non-shift model, the density estimation is not necessary, which is an attractive feature in small to moderate sized sample case.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.1
• /
• pp.125-133
• /
• 2004

In this paper, we present a new method that combines the characteristics of edge in-formation and second-order neural networks for the classification of structural textures. The edges of a texture are extracted using an edge detection approach. From this edge information, classification features called second-order features are obtained. These features are fed into a second-order neural network for training and subsequent classification. It will be shown that the main disadvantage of using structural methods in texture classifications, namely, the difficulty of the extraction of texels, is overcome by the proposed method.

• Journal of the Korean Data and Information Science Society
• /
• v.4
• /
• pp.31-44
• /
• 1993

Multivariate exponentially weighted moving average (EWMA) control charts for monitoring the variance-covariance matrix are investigated. A variable sampling interval (VSI) feature is considered in these charts. Multivariate EWMA control charts for monitoring the variance-covariance matrix are compared on the basis of their average time to signal (ATS) performances. The numerical results show that multivariate VSI EWMA control charts are more efficient than corrsponding multivariate fixed sampling interval (FSI) EWMA control charts.

• Communications for Statistical Applications and Methods
• /
• v.19 no.5
• /
• pp.655-662
• /
• 2012

Classification is an important research field in pattern recognition with high-dimensional predictors. The support vector machine(SVM) is a penalized feature selector and classifier. It is based on the hinge loss function, the non-convex penalty function, and the smoothly clipped absolute deviation(SCAD) suggested by Fan and Li (2001). We developed the algorithm for the multiclass SVM with the SCAD penalty function using the local quadratic approximation. For multiclass problems we compared the performance of the SVM with the $L_1$, $L_2$ penalty functions and the developed method.