Theory and Applications of a Nonlinear Principal Component Analysis

摘要

Principal Component Analysis (PCA) has been applied in various areas such as pattern recognition and data compression. In some cases, however, PCA does not extract the characteristics of the data-distribution efficiently. In order to overcome this problem, we have proposed a novel method of nonlinear PCA (NLPCA) which preserves the order of the principal components and we have implemented the NLPCA with neural networks. In this paper, we discuss the property of the proposed NLPCA in regard with a curvilinear axis and a contour map with some simulated results.