• Proceedings of the Korean Environmental Sciences Society Conference
• /
• 2003.11a
• /
• pp.33-36
• /
• 2003

Principal component analysis (PCA) was used to improve image classification by the unsupervised classification techniques, the K-means. To do this, I selected a Landsat TM scene of Jeju Island, Korea and proposed two methods for PCA： unstandardized PCA (UPCA) and standardized PCA (SPCA). The estimated accuracy of the image classification of Jeju area was computed by error matrix. The error matrix was derived from three unsupervised classification methods. Error matrices indicated that classifications done on the first three principal components for UPCA and SPCA of the scene were more accurate than those done on the seven bands of TM data and that also the results of UPCA and SPCA were better than those of the raw Landsat TM data. The classification of TM data by the K-means algorithm was particularly poor at distinguishing different land covers on the island. From the classification results, we also found that the principal component based classifications had characteristics independent of the unsupervised techniques (numerical algorithms) while the TM data based classifications were very dependent upon the techniques. This means that PCA data has uniform characteristics for image classification that are less affected by choice of classification scheme. In the results, we also found that UPCA results are better than SPCA since UPCA has wider range of digital number of an image.

• Journal of Korean Society for Atmospheric Environment
• /
• v.18 no.1
• /
• pp.25-37
• /
• 2002

The purpose of the study was to survey the distribution patterns of inorganic elements of PM-10 in the various indoor environments and analyze the pollution patterns of aerosol in various places of indoor environment using a pattern recognition method based on cluster analysis and disjoint principal component analysis. A total of 40 samples in the indoor had been collected using mini-vol portable samplers. These samples were analyzed for their 19 bulk inorganic compounds such as B, Na, Mg, Al, K, Ca, Ti, V, Cr, Fe, Ni, Cu, Zn, As, Se, Cd, Ba, Ce, and Pb by using an ICP-MS. By applying a disjoint principal component analysis, four patterns of the indoor air pollutions were distinguished. The first pattern was identified as a group with high concentrations of PM-10, Na, Mg, and Ca. The second pattern was identified as a group with high concentrations B, Mg, At, Ca, Fe, Cu, and Ba. The third pattern was a group of sites with high concentrations of K, Zn. Cd. The fourth pattern was a group with low concentrations PM-10 and all inorganic elements. This methodology was found to be helpful enough to set the criteria standard of indoor air quality, corresponding pollutants, and classification of indoor environment categories when making an indoor air quality law.

• Journal of Biomedical Engineering Research
• /
• v.27 no.2
• /
• pp.59-63
• /
• 2006

Magnetocardiogram (MCG) topography is a useful diagnostic technique that employs multi-channel magnetocardiograms. Measurement of artifact-free MCG signals is essenctial to obtain MCG topography or map for a diagnosis of human heart. Principal component analysis (PCA) combined with an artificial neural network (ANN) is proposed to remove a pulse-type artifact in the MCG signals. The algorithm is composed of a PCA module which decomposes the obtained signal into its principal components, followed by an ANN module for the classification of the components automatically. In the experiments with volunteer subjects, 97% of the decisions that were made by the ANN were identical to those by the human experts. Using the proposed technique, the MCG topography was successfully obtained without the artifact.

• Journal of the Korea Safety Management & Science
• /
• v.12 no.2
• /
• pp.231-236
• /
• 2010

The investigation was initiated with data from 27 abandoned mines along with 12 locations in Kyongbuk abandoned mines. The analyses for soil pollution by heavy metal pollutants were conducted by using correlation analysis, cluster analysis, and principal component analysis. The correlation analysis indicated that Ni and pH were highly correlated compared to those of other heavy metal ions. The principal component analyses showed that the heavy metal ions might be classified into two catagories, such as antropogenic and lithogenic components. The cluster analysis was also clearly divided by two groups. The respective two groups might be Pb-Zn-Cd-Cu and As-Hg-Ni.

• Journal of Korea Water Resources Association
• /
• v.38 no.12 s.161
• /
• pp.985-994
• /
• 2005

We examined problems of the principal component analysis(PCA), which is able to analyze at the low dimensionality as a methodologv to assess hydrologic time series, and introduced the theory and characteristics of independent component analysis(ICA) that can supplement problems of principal component analysis. We also applied the global sea surface temperature(SST) of the Nino region and assessed the correlation between El $\tilde{n}ino$-Southern Oscillation(ENSO) and SST. The results of examining separation-ability of principal components using mixed signals indicate that the independent component analysis is statistically superior compared to that of the principal component analysis. Finally, we assessed correlation between ENSO and global anomaly SST. The independent component analysis was applied to the $5^{\circ}{\times}5^{\circ}$(latitude and longitude) global anomaly SST in the Nino+3.4 region that is the El $\tilde{n}ino$ observation section. We assessed the correlation with the ENSO years. These results of the analysis show that only one independent component($86\%$) was able to represent the entire behavior and was consistent with the main ENSO years. Finally, we carried out independent component analysis for summer seasonal rainfalls at nine stations and could extract ICs to reflect geographical characteristics. The increasing trend has been shown at IC-1 and IC-2 since 1970s.

• The Korean Journal of Applied Statistics
• /
• v.14 no.1
• /
• pp.137-148
• /
• 2001

Compositional data found in many practical applications consist of non-negative vectors of proportions with the constraint which the sum of the elements of each vector is unity. It is well-known that the statistical analysis of compositional data suffers from the unit-sum constraint. Moreover, the non-linear pattern frequently displayed by the data does not facilitate the application of the linear multivariate techniques such as principal component analysis. In this paper we develop new type of principal component analysis for compositional data using Box-Cox contrast transformation. Numerical illustrations are provided for comparative purpose.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.1
• /
• pp.143-151
• /
• 2015

In this paper, we investigate how the power consumption of a heat pump dryer depends on various factors in the drying process by analyzing variables that affect the power consumption. Since there are in general many variables that affect the power consumption, for a feasible analysis, we utilize the principal component analysis to reduce the number of variables (or dimensionality) to two or three. We find that the first component is correlated positively to the entrance temperature of various devices such as compressor, expander, evaporator, and the second, negatively to condenser. We then model the power consumption as a multiple regression with two and/or three transformed variables of the selected principal components. We find that fitted value from the multiple regression explains 80~90% of the observed value of the power consumption. This results can be applied to a more elaborate control of the power consumption in the heat pump dryer.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.4
• /
• pp.721-731
• /
• 2017

In this study, principal component analysis methods using Lasso penalty are introduced. There are two popular methods that apply Lasso penalty to principal component analysis. The first method is to find an optimal vector of linear combination as the regression coefficient vector of regressing for each principal component on the original data matrix with Lasso penalty (elastic net penalty in general). The second method is to find an optimal vector of linear combination by minimizing the residual matrix obtained from approximating the original matrix by the singular value decomposition with Lasso penalty. In this study, we have reviewed two methods of principal components using Lasso penalty in detail, and shown that these methods have an advantage especially in applying to data sets that have more variables than cases. Also, these methods are compared in an application to a real data set using R program. More specifically, these methods are applied to the crime data in Ahamad (1967), which has more variables than cases.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2012.10a
• /
• pp.241-242
• /
• 2012

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.5
• /
• pp.7-12
• /
• 2011

In this papwer, we propose a new feature extraction method, named as Class-augmented Kernel Principal Component Analysis (CA-KPCA), which can extract nonlinear features for classification. Among the subspace method that was being widely used for feature extraction, Class-augmented Principal Component Analysis (CA-PCA) is a recently one that can extract features for a accurate classification without computational difficulties of other methods such as Linear Discriminant Analysis (LDA). However, the features extracted by CA-PCA is still restricted to be in a linear subspace of the original data space, which limites the use of this method for various problems requiring nonlinear features. To resolve this limitation, we apply a kernel trick to develop a new version of CA-PCA to extract nonlinear features, and evaluate its performance by experiments using data sets in the UCI Machine Learning Repository.