• Journal of the Korea Society of Computer and Information
• /
• v.16 no.4
• /
• pp.15-22
• /
• 2011

Principal component analysis (PCA) is a well-known method for dimensionality reduction and feature extraction while maintaining most of the variation in data. Although PCA has been applied in many areas successfully, it is sensitive to outliers and only valid for Gaussian distributions. Several variants of PCA have been proposed to resolve noise sensitivity and, among the variants, improved robust fuzzy PCA (RF-PCA2) demonstrated promising results. RF-PCA, however, is still a linear algorithm that cannot accommodate non-Gaussian distributions. In this paper, a non-linear algorithm that combines RF-PCA2 and kernel PCA (K-PCA), called improved robust kernel fuzzy PCA (RKF-PCA2), is introduced. The kernel methods make it to accommodate non-Gaussian distributions. RKF-PCA2 inherits noise robustness from RF-PCA2 and non-linearity from K-PCA. RKF-PCA2 outperforms previous methods in handling non-Gaussian distributions in a noise robust way. Experimental results also support this.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.5
• /
• pp.1093-1102
• /
• 2010

Principal component analysis (PCA) is a well-known method for dimension reduction while maintaining most of the variation in data. Although PCA has been applied to many areas successfully, it is sensitive to outliers. Several variants of PCA have been proposed to resolve the problem and, among the variants, robust fuzzy PCA (RF-PCA) demonstrated promising results. RF-PCA uses fuzzy memberships to reduce the noise sensitivity. However, there are also problems in RF-PCA and the convergence property is one of them. RF-PCA uses two different objective functions to update memberships and principal components, which is the main reason of the lack of convergence property. The difference between two functions also slows the convergence and deteriorates the solutions of RF-PCA. In this paper, a variant of RF-PCA, called RF-PCA2, is proposed. RF-PCA2 uses an integrated objective function both for memberships and principal components. By using alternating optimization, RF-PCA2 is guaranteed to converge on a local optimum. Furthermore, RF-PCA2 converges faster than RF-PCA and the solutions found are more similar to the desired solutions than those of RF-PCA. Experimental results also support this.

• The Korean Journal of Applied Statistics
• /
• v.30 no.3
• /
• pp.441-451
• /
• 2017

In this paper, we review principal component analysis (PCA) procedures in the frequency domain and apply them to analyze sea surface temperature data. The classical PCA defined in the time domain is a popular dimension reduction technique. Extending the conventional PCA to the frequency domain makes it possible to define PCA in the frequency domain, which is useful for dimension reduction as well as a feature extraction of multiple time series. We focus on two PCA methods in the frequency domain, Hilbert PCA (HPCA) and frequency domain PCA (FDPCA). We review these two PCAs in order for potential readers to easily understand insights as well as perform a numerical study for comparison with conventional PCA. Furthermore, we apply PCA methods in the frequency domain to sea surface temperature data on the tropical Pacific Ocean. Results from numerical experiments demonstrate that PCA in the frequency domain is effective for the analysis of time series data.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.963-977
• /
• 2011

PCA(Principal Component Analysis) is a well-studied statistical technique and an important tool for handling multivariate data. Although many algorithms exist for PCA, most of them are unsuitable for real time applications or high dimensional problems. Since it is desirable to avoid extensive matrix operations in such cases, alternative solutions are required to calculate the eigenvalues and eigenvectors of the sample covariance matrix. Erdogmus et al. (2004) proposed Recursive PCA(RPCA), which is a fast adaptive on-line solution for PCA, based on the first order perturbation theory. It facilitates the real-time implementation of PCA by recursively approximating updated eigenvalues and eigenvectors. However, the performance of the RPCA method becomes questionable as the size of newly-added data increases. In this paper, we modified the RPCA method by taking advantage of the mathematical relation of eigenvalues and eigenvectors of sample covariance matrix. We compared the performance of the proposed algorithm with that of RPCA, and found that the accuracy of the proposed method remarkably improved.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.146-148
• /
• 1999

문서분류의 문제점 중의 하나는 사용하는 데이터의 차원이 매우 크다는 것이다. 그러므로 문서에서 필요한 단어만을 자동적으로 추출하여 문서데이터의 차원을 축소하는 작업이 문서분류에서는 필수적이다. DF(Document Frequency)는 문서의 차원축소의 대표적인 통계적 방법 중 하나인데, 본 논문에서는 문서의 차원축소에 DF와 주성분 분석(PCA)을 비교하여 주성분 분석이 문서의 차원축소에 적합함을 실험적으로 보인다. 그리고 비선형 주성분 분석(nonlinear PCA) 방법 중 locally linear PCA와 kenel PCA를 적용하여 비선형 주성분 분석을 이용하여 문서의 차원을 줄이는 것이 선형 주성분 분석을 이용하는 것 보다 문서분류에 더 적합함을 실험적으로 보인다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.6 no.3
• /
• pp.249-254
• /
• 2005

In this paper, we analysis the recognition performance of PCA/LDA by distance measures. We are adapt to ORL face database with the fourteen distance measures. In case of PCA, it has high performance for the manhattan distance and the weighted SSE distance to face recognition, In case of PCA/LDA, it has high performance for the angle-based distance and the modified SSE distance. Also, PCA/LDA is better than PCA for reduction of dimension. Therefore, the PCA/LDA method and the angle-based distance have the most performance and a few dimension for face recognition with ORL face database.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.959-971
• /
• 2019

CNN (Convolutional Neural Network) is one representative deep learning algorithm, which can extract high-level spatial and spectral features, and has been applied for hyperspectral image classification. However, one significant drawback behind the application of CNNs in hyperspectral images is the high dimensionality of the data, which increases the training time and processing complexity. To address this problem, several CNN based hyperspectral image classification studies have exploited PCA (Principal Component Analysis) for dimensionality reduction. One limitation to this is that the spectral information of the original image can be lost through PCA. Although it is clear that the use of PCA affects the accuracy and the CNN training time, the impact of PCA for CNN based hyperspectral image classification has been understudied. The purpose of this study is to analyze the quantitative effect of PCA in CNN for hyperspectral image classification. The hyperspectral images were first transformed through PCA and applied into the CNN model by varying the size of the reduced dimensionality. In addition, 2D-CNN and 3D-CNN frameworks were applied to analyze the sensitivity of the PCA with respect to the convolution kernel in the model. Experimental results were evaluated based on classification accuracy, learning time, variance ratio, and training process. The size of the reduced dimensionality was the most efficient when the explained variance ratio recorded 99.7%~99.8%. Since the 3D kernel had higher classification accuracy in the original-CNN than the PCA-CNN in comparison to the 2D-CNN, the results revealed that the dimensionality reduction was relatively less effective in 3D kernel.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.848-850
• /
• 2007

본 논문은 치아의 모양, 크기 및 턱의 모양 등과 같은 치아의 기하학적인 특징들을 사용하여 치아의 성 변별시스템에 PCA기법과 LDA기법을 각각 적용하고 두 기법을 비교분석한다. PCA기법과 LDA기법은 생체인식을 위한 주요 매핑기법으로 알려져 있다. PCA분석 기법을 적용하여 성변별의 결과 76%의 인식률이 획득되었으며, LDA분석기법은 66%의 인식률이 획득되었다. 본 연구의 결과로부터 PCA기법은 치아의 성변별에 있어 LDA기법보다 우수한 성능을 제공함을 확인할 수 있었다.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.2
• /
• pp.25-31
• /
• 2011

Principal component analysis (PCA) is a well-known method for dimensionality reduction and feature extraction. Although PCA has been applied in many areas successfully, it is sensitive to outliers due to the use of sum-square-error. Several variants of PCA have been proposed to resolve the noise sensitivity and, among the variants, improved robust fuzzy PCA (RF-PCA2) demonstrated promising results. RF-PCA2, however, still can fall into a local optimum due to equal initial membership values for all data points. Another reason comes from the fact that RF-PCA2 is based on sum-square-error although fuzzy memberships are incorporated. In this paper, a variant of RF-PCA2 called RF-PCA3 is proposed. The proposed algorithm is based on the objective function of RF-PCA2. RF-PCA3 augments RF-PCA2 with the objective function of PCA and initial membership calculation using data distribution, which make RF-PCA3 to have more chance to converge on a better solution than that of RF-PCA2. RF-PCA3 outperforms RF-PCA2, which is demonstrated by experimental results.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.6
• /
• pp.452-456
• /
• 2003

We discuss how to reduce the number of inverse matrix and its dimensions requested in MLLR framework for speaker adaptation. To find a smaller set of variables with less redundancy, we adapt PCA (principal component analysis) and ICA (independent component analysis) that would give as good a representation as possible. The amount of additional computation when PCA or ICA is applied is as small as it can be disregarded. 10 components for ICA and 12 components for PCA represent similar performance with 36 components for ordinary MLLR framework. If dimension of SI model parameter is n, the amount of computation of inverse matrix in MLLR is proportioned to O(n⁴). So, compared with ordinary MLLR, the amount of total computation requested in speaker adaptation is reduced by about 1/81 in MLLR with PCA and 1/167 in MLLR with ICA.