• Communications for Statistical Applications and Methods
• /
• v.22 no.2
• /
• pp.173-180
• /
• 2015

We propose a selection procedure of principal components in principal component regression. Our method selects principal components using variable selection procedures instead of a small subset of major principal components in principal component regression. Our procedure consists of two steps to improve estimation and prediction. First, we reduce the number of principal components using the conventional principal component regression to yield the set of candidate principal components and then select principal components among the candidate set using sparse regression techniques. The performance of our proposals is demonstrated numerically and compared with the typical dimension reduction approaches (including principal component regression and partial least square regression) using synthetic and real datasets.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.05a
• /
• pp.151-156
• /
• 2003

Variable selection algorithm for principal component analysis using penalized likelihood method is proposed. We will adopt a probabilistic principal component idea to utilize likelihood function for the problem and use HARD penalty function to force coefficients of any irrelevant variables for each component to zero. Consistency and sparsity of coefficient estimates will be provided with results of small simulated and illustrative real examples.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.12
• /
• pp.2511-2519
• /
• 2009

In this paper, we introduce the methodological system design via feature selection using Principal Component Analysis and Particle Swarm Optimization algorithms. The overall methodological system design comes from three kinds of modules such as preprocessing module, feature extraction module, and recognition module. First, Histogram equalization enhance the quality of image by exploiting contrast effect based on the normalized function generated from histogram distribution values of 2D face image. Secondly, PCA extracts feature vectors to be used for face recognition by using eigenvalues and eigenvectors obtained from covariance matrix. Finally the feature selection for face recognition among the entire feature vectors is considered by means of the Particle Swarm Optimization. The optimized Polynomial-based Radial Basis Function Neural Networks are used to evaluate the face recognition performance. This study shows that the proposed methodological system design is effective to the analysis of preferred face recognition.

• Journal of Korean Institute of Industrial Engineers
• /
• v.40 no.3
• /
• pp.275-282
• /
• 2014

One of the most widely used methods for dimensionality reduction is principal component analysis (PCA). However, the reduced dimensions from PCA do not provide a clear interpretation with respect to the original features because they are linear combinations of a large number of original features. This interpretation problem can be overcome by feature selection approaches that identifying the best subset of given features. In this study, we propose an unsupervised feature selection method based on the geometrical information of PCA loading vectors. Experimental results from a simulation study demonstrated the efficiency and usefulness of the proposed method.

• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.1025-1036
• /
• 2003

Variable selection algorithm for principal component analysis using penalty function is proposed. We use the fact that usual principal component problem can be expressed as a maximization problem with appropriate constraints and we will add penalty function to this maximization problem. Simulated annealing algorithm is used in searching for optimal solutions with penalty functions. Comparisons between several well-known penalty functions through simulation reveals that the HARD penalty function should be suggested as the best one in several aspects. Illustrations with real and simulated examples are provided.

• Journal of Pharmaceutical Investigation
• /
• v.37 no.6
• /
• pp.377-395
• /
• 2007

Near infrared(NIR) spectral data have been used for the noninvasive analysis of various biological samples. Nonetheless, absorption bands of NIR region are overlapped extensively. It is very difficult to select the proper wavelengths of spectral data, which give the best PCR(principal component regression) models for the analysis of constituents of biological samples. The NIR data were used after polynomial smoothing and differentiation of 1st order, using Savitzky-Golay filters. To find the best PCR models, all-possible combinations of available principal components from the given NIR spectral data were derived by in-house programs written in MATLAB codes. All of the extensively generated PCR models were compared in terms of SEC(standard error of calibration), $R^2$, SEP(standard error of prediction) and SECP(standard error of calibration and prediction) to find the best combination of principal components of the initial PCR models. The initial PCR models were found by SEC or Malinowski's indicator function and a priori selection of spectral points were examined in terms of correlation coefficients between NIR data at each wavelength and corresponding concentrations. For the test of the developed program, aqueous solutions of BSA(bovine serum albumin) and glucose were prepared and analyzed. As a result, the best PCR models were found using a priori selection of spectral points and the final model selection by SEP or SECP.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.5
• /
• pp.1013-1022
• /
• 2014

Correlation Power Analysis (CPA) is a type of Side-Channel Analysis (SCA) that extracts the secret key using the correlation coefficient both side-channel information leakage by cryptography device and intermediate value of algorithms. Attack performance of the CPA is affected by noise and temporal synchronization of power consumption leaked. In the recent years, various researches about the signal processing have been presented to improve the performance of power analysis. Among these signal processing techniques, compression techniques of the signal based on Principal Component Analysis (PCA) has been presented. Selection of the principal components is an important issue in signal compression based on PCA. Because selection of the principal component will affect the performance of the analysis. In this paper, we present a method of selecting the principal component by using the correlation of the principal components and the power consumption is high and a CPA technique based on the principal component that utilizes the feature that the principal component has different. Also, we prove the performance of our method by carrying out the experiment.

• Korean Journal of Veterinary Research
• /
• v.53 no.1
• /
• pp.7-10
• /
• 2013

A study was conducted to determine the interdependence among the conformation traits of 28 "Pallaresa" cows using principal component analysis. Originally 21 body linear measurements were obtained, from which eight traits are subsequently eliminated. From the principal components analysis, with raw varimax rotation of the transformation matrix, two principal components were extracted, which accounted for 65.8% of the total variance. The first principal component alone explained 51.6% of the variation, and tended to describe general size, while the second principal component had its loadings for back-sternal diameter. The two extracted principal components, which are traits related to dorsal heights and back-sternal diameter, could be considered in selection programs.

• Journal of applied mathematics & informatics
• /
• v.14 no.1_2
• /
• pp.377-386
• /
• 2004

A technique that provides prediction intervals based on a model called an empirical linear model is discussed. The technique, high-dimensional empirical linear prediction (HELP), involves principal component analysis, factor analysis and model selection. HELP can be viewed as a technique that provides prediction (and confidence) intervals based on a factor analysis models do not typically have justifiable theory due to nonidentifiability, we show that the intervals are justifiable asymptotically.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.175-178
• /
• 2006

본 논문에서는 주요성분분석과 상호정보 추정을 조합한 입력변수선택 기법을 제안하였다. 여기서 주요성분분석은 2차원 통계성을 이용하여 입력변수 간의 독립성을 찾기 위함이고, 상호정보의 추정은 적응적 분할을 이용하여 입력변수의 확률밀도함수를 계산함으로써 변수상호간의 종속성을 좀더 정확하게 측정하기 위함이다. 제안된 기법을 인위적으로 제시된 각 500개의 샘플을 가지는 6개의 독립신호와 1개의 종속신호를 대상으로 실험한 결과, 빠르고 정확한 변수의 선택이 이루어짐을 확인하였다.