论文信息 - A Comparative Study of Principal Component Analysis Techniques

A Comparative Study of Principal Component Analysis Techniques

Principal Component Analysis (PCA) is a useful technique for reducing the dimensionality of datasets for compression or recognition purposes. Many different methods have been proposed for performing PCA. This study aims to compare these methods by analysing the solutions which these methods find. We have estimated the correlation between these solutions and produced the errors using bootstrap resampling.

M. Jabri | M. Partridge | R. Calvo

[1] Rafael A. Calvo,et al. Fast Dimensionality Reduction and Simple PCA , 1998, Intell. Data Anal..

[2] Rafael A. Calvo,et al. High Performance Principal Component Analysis with Paral , 1998 .

[3] E. Smith. Methods of Multivariate Analysis , 1997 .

[4] J. E. Jackson,et al. Statistical Factor Analysis and Related Methods: Theory and Applications , 1995 .

[5] Yann LeCun,et al. Efficient Pattern Recognition Using a New Transformation Distance , 1992, NIPS.

[6] F. A. Seiler,et al. Numerical Recipes in C: The Art of Scientific Computing , 1989 .

[7] George Henry Dunteman,et al. Introduction To Multivariate Analysis , 1984 .

[8] R. Frisch. Correlation and scatter in statistical variables , 1951 .

[9] H. Hotelling. Analysis of a complex of statistical variables into principal components. , 1933 .